An ontology is essentially a shared data model that simplifies the process of connecting applications in a particular domain, and it’s one of the core elements for developing digital twins.

“Microsoft is investing heavily in enabling our partners with the technology and services they need to create digital twin solutions that support new and existing needs of the world’s largest real estate portfolios,” said Microsoft Azure IoT general manager Tony Shakib.

This recent push into construction extends the utility of Microsoft’s Azure Digital Twins, released last year.

To gain a foothold in the field, Microsoft partnered with RealEstateCore, a Swedish consortium of real estate owners, tech companies, and research groups, to integrate these services with various industry standards. Making a Smart Building RealEstateCore ontology for Azure Digital Twins enables the various parties in building markets — owners, construction teams, and vendors — to collaborate and communicate about real estate.

This could accelerate the ability to weave IoT data, AI models, and analytics into digital twins, and to help simplify the transition to sustainable and green innovation, currently one of one of the fastest-growing venture capital sectors.

Accelerating digital transformation

Digital transformation has been slow to develop in construction and real estate markets. Microsoft believes that the development of better standards and integrations could help accelerate such transformation. That is important if only because real estate represents one of the largest asset classes in the world. In its recent Global Building Stock Database update, Guidehouse Insights predicts the square footage of buildings will grow from about 166 billion square meters in 2020 to 196 billion square meters in 2030.

Building owners are hoping that digital twins could help increase the value of their existing holdings at less cost than building new ones.

But figuring out how to increase building asset value and net operating income is a complicated problem that spans technology and change management issues, Shakib said.

This shift is further complicated by challenges in retrofitting digital twins’ capabilities to existing building management systems. Shakib said many building management and automation vendors have attempted to limit buildings to custom, proprietary “walled garden” approaches that can hurt clients in the long run.

Better ontologies could smooth this transition. Such thinking was behind the RealEstateCore Consortium, which was born out of a partnership between academia and industry. The consortium created the RealEstateCore ontology that employed a graph data model and built on years of best practices gleaned from experience with larger property owners such as Vasakronan.

RealEstateCore can provide a bridge to various building industry standards such as Brick Schema, Project Haystack, W3C Building Topology Ontology (W3C BOT), and more. Today, different partners can run into problems integrating applications using custom data formats. This is especially relevant in construction, as there are huge pitfalls from data loss in the steps from building design to construction, commission, handover, and operation.

Seeing a return

Improved digital twins promise significant ROI for building owners and operators. By improving the categorization, integration, and fidelity of data, digital twin developers can create better digital replicas of physical buildings and the components they comprise.

Some of the early gains come from cost savings related to energy efficiency. Microsoft has been exploring these techniques on its campuses to realize 20% to 30% energy savings. These projects can start by harvesting data from existing building control systems to find room for improvement.

Microsoft’s Project Bonsai has been able to squeeze an additional 10% to 15% of savings by applying AI to optimize controls further. Down the road, the U.S. Department of Energy’s Grid-Interactive Efficient Buildings could help owners save even more by enabling their facilities to interact with the digital electric grid in real time.

Beyond energy savings, there has been rapidly growing interest in using digital twins to optimize building space, activate building amenities, and support various health and wellness scenarios in the wake of COVID. For example, RXR Realty uses Azure Digital Twins to combine building data with people counting, social distance detection, face mask detection, and air quality monitoring to provide a building wellness index. The appropriate ontology also allowed them to capture important metrics while still respecting privacy and ethics.

Turning things into assets

Digital twins help a group of people make sense of the data surfaced by IoT devices. An ontology provides a set of models for wiring these up in a particular domain, such as a building structure, system, city, or energy grid.

An ontology can provide a starting point for organizing the information to solve a problem that spans different roles, such as designers, builders, vendors, and operators. For example, a construction team might need to know how to install a new heater; a general contractor would want to know how long installing it will take, while the owner would want to know the appropriate maintenance schedule.

The built world is complex, and a smart building’s ontology must seek to represent that intricate reality in a way that is simple for developers to use. “An ontology must balance power and comprehensiveness with simplicity and ease of use to generate enough adoption,” Shakib said.

All of the major cloud vendors have announced various kinds of IoT initiatives for helping to weave sensors and actuators into new cloud applications. But Microsoft has been the only one to champion digital twins thus far. The real value of digital twins lies in helping decision-makers frame how their decisions about these IoT-related applications can be woven together to impact assets in the real world.

The justices, in a 6-3 decision authored by conservative Justice Amy Coney Barrett, sided with former Cumming, Georgia police sergeant Nathan Van Buren in an appeal of his conviction under the Computer Fraud and Abuse Act, reversing a lower court ruling that had upheld a jury verdict against him.

Van Buren was charged after a 2015 FBI sting operation. The Supreme Court concluded that Van Buren could not be convicted for misusing the database to perform the investigation because the information had been available to him as part of his job.

“This provision covers those who obtain information from particular areas in the computer — such as files, folders or databases — to which their computer access does not extend. It does not cover those who, like Van Buren, have improper motives for obtaining information that is otherwise available to them,” Barrett wrote in the ruling.

Three of the court’s conservative justices, Clarence Thomas, John Roberts and Samuel Alito, dissented from the ruling.

Van Buren’s attorney, Jeffrey Fisher, welcomed the ruling and said the computer fraud law “is now restricted to its proper reach.” The U.S. Justice Department, which had defended the conviction, declined to comment.

The dispute centered on a 1986 U.S. law meant to target hacking and related computer crimes. The law prohibits accessing a computer without authorization and also exceeding authorized access.

At issue was whether a person with authority to access a computer can be guilty of fraud for then misusing it. Van Buren had argued that the law targets only those with no right whatsoever to access the computer, whereas he was entitled to use the law enforcement database, even if it was for an inappropriate reason.

The Justice Department argued that the law “aims directly at ‘insider’ conduct” like Van Buren’s “forbidden use of his law enforcement credentials.”

The dissenting justices had a different interpretation of the law’s text, finding that Van Buren’s actions were illegal because he was not entitled to the information he was otherwise authorized to access.

“Using a police database to obtain information in circumstances where that use is expressly forbidden is a crime,” Thomas wrote in his dissent.

Suffering financial difficulties, Van Buren had asked a local man, Andrew Albo, for money. Albo alerted law enforcement authorities and the FBI devised a sting in which Albo offered to pay Van Buren money to run a search for a license plate on a law enforcement database. Albo’s fictional story was that he wanted to find out if a local stripper was an undercover cop.

Albo gave Van Buren $6,000 and Van Buren conducted the search. A jury convicted Van Buren in 2017 of violating the computer fraud law and a separate count of honest services fraud. The Atlanta-based 11th U.S. Circuit Court of Appeals in 2019 upheld the computer fraud conviction, but ordered a retrial on the other charge.

Barrett noted in the ruling that an overly broad view of the law would penalize commonplace computer activity such as an employee sending a personal email or checking sports scores on a work device. If the law “criminalizes every violation of a computer-use policy, then millions of otherwise law-abiding citizens are criminals,” Barrett wrote.

Former President Donald Trump’s three conservative appointees to the court — Barrett, Brett Kavanaugh and Neil Gorsuch — were joined by the three liberal justices in the ruling.

Istanbul-based Veloxia is the latest Turkish game company to raise a significant round of money. In mobile games, Turkey is starting to become so hot that you could compare it to Finland, which gave birth to gaming companies such as Supercell and Rovio. In recent years, Turkish game companies have also become hot commodities, with Zynga acquiring Peak Games last year for $1.8 billion.

Collective Spark led the round with participation from Bogazici Ventures, Ak Portfoy Girisim Sermayesi, and several angel investors. Idle games are titles that you play when you have a lot of idle time, and they’re often quite repetitive, where you tap the screen repeatedly.

Origins

Above: Veloxia’s founders have raised $4 million to date.

Tugay Alyıldız, Melih Gündoğdu, and Mehveş Altay started the company in June 2019 with the mission of creating innovative products on a global scale in mobile gaming. Veloxia published its first product, an idle game called “Space Colony: Idle” on iOS and Android platforms in late 2019, and it has been ranked among the top strategy games and achieved nearly a million downloads.

The trio started the company to create multiplayer games that can spread in a viral fashion with more efficient marketing and high profitability. The company wants to build a portfolio of games across a variety of genres. Alyildiz has experience in consulting, strategy, investment, and finance. Gündoğdu, chief operating officer, is a management engineering graduate, with experience in consulting, strategy, investment finance. And Altay is the chief technology officer.

“Our infrastructure is backed by behavioral economy models which we are constantly developing as we scale our products,” said Tugay Alyildiz, CEO of Veloxia, in an email to GamesBeat. “We work towards differentiating Veloxia from other mobile gaming startups with our approach to monetization in game design and by successfully managing products in different categories under a cluster structure.”

He said the funding round will help the company put its idle know-how to work, further strengthen its analytics muscle, and expand across strategy and action game categories.

Alyildiz said that the development team has been growing under the leadership of Altay, who worked for Microsoft and Google in several countries during her college years. She said the company has an international team with 22 members living in different geographies. The company will expand in Turkey and aboard and maintain its inclusive and diverse team culture.

Big ambitions

Above: Veloxia’s Space Colony: Idle game.

Image Credit: Veloxia

Görkem Oktay, general partner at Collective Spark, said in a statement that Turkey has been at the forefront of mobile gaming with many successful startups and multiple exits in recent years. In this context, he said his firm has been closely following the landscape for a long time and invested in the Veloxia team which he has known since inception. He said he looks forward to Veloxia publishing new games in various categories.

Alyıldız noted that he will be attending the Game Developers Conference in San Francisco this year on behalf of Veloxia to help raise awareness of Veloxia in the mobile game community.

The company is currently working on a title in the multiplayer online battle arena games (MOBA) category. The hope is to produce the biggest game out of Turkey and to expand its overseas team and open North Europe offices. The company has raised almost $4 million to date.

Streamlytics, a Miami-based data provider founded in 2018, believes letting users sell their data could be part of the solution. The company has already collected more than 75 million data points this way and says it aims to “democratize” data by giving users more control and then selling the user-supplied data to enterprises — from media conglomerates to consumer goods companies. Streamlytics is particularly focused on working with Black users in the U.S. and on getting underrepresented data into AI training models.

To begin, people download their data from whatever platforms they choose and then upload it to Streamlytics’ own consumer application, Clture. Users sign a data license that says they own their data (rather than Streamlytics or the companies that will later pay for it), request a payout, and get paid. The company’s patent-pending data standard then unifies the various file types and data sources, packaging it up nicely for enterprise customers. Companies can only buy aggregate data, either via a data feed, a specific number of data points, or another offering customized for their use cases or specific needs. Users currently do not know who is purchasing their data and cannot opt out of specific companies, but Streamlytics is thinking about adding this capability in the future.

Streamlytics founder and CEO Angela Benton believes this approach is necessary not just to begin compensating users more equitably, but to make sure companies are using better, less biased data to build their technologies. She spoke with VentureBeat about the changing data privacy landscape and how Streamlytics wants to shift the status quo.

This interview has been edited for brevity and clarity.

VentureBeat: What does it mean to democratize data?

Angela Benton: For me, it’s less about democratizing for companies because that pretty much exists right now. But consumers are just finally starting to really understand that they’re creating data, how much of it they’re creating, and that they don’t really own it or have any agency over it. I personally tend to think about it in terms of how much money is being made (both in terms of revenue and companies leveraging data internally), and I feel like consumers should get some kind of compensation. People think of it as a negative — that these companies are leveraging data. But to me, the real issue is that there’s no equality in the relationship between the people who are creating the data and the people who are actually leveraging it. This is about leveling the playing field.

VentureBeat: And so why do we need to do this, especially in terms of enterprises? With people becoming more aware, there’s of course the issue of customer trust. But how might this approach to data be good for enterprises, too?

Benton: When I talk about the importance of consumers having agency over their data, people tend to think of advertising and marketing. I actually think about it in terms of AI. Everything is going to be powered by AI. You can see its growth over not even the last 10 or five years, but the past two years. And so if the algorithms within the AI ecosystem aren’t trained on data that is diverse, that correctly represents the gender and ethnic makeup of the world we live in today, that for me is the bigger problem. That’s how you end up with bias. To me, that’s the bigger implication of why data is so important.

VentureBeat: So it’s not even just a question of if this current system is fair, but also getting better data. How can this change how enterprises acquire and use data while also improving the data itself?

Benton: I’ll give you an example, and this is actually interesting because it applies to what happened in 2020 with the Black Lives Matter movement and everything that happened with George Floyd. You have a lot of advertisers who are interested in, I don’t necessarily want to say “targeting” the African African American community, but they want to target them in a way where they’re doing some kind of social good. And with the changes to third-party tracking, they don’t really have a way to understand who is, for instance, African American, and who is not.

We’re working with a large media company that has brands coming to it to do that. We’re the largest first-party data provider for African American data that is sourced in this ethical way, and we can make recommendations based on activity that falls kind of within a demographic. So think about how you typically stream — you’re probably also looking at what you’re going to order on UberEats, shopping on Amazon, and more. So if they want to understand African American women ages 18 to 24, what our data uniquely does is say they watch Bridgerton on Netflix, but also maps to other things they’re doing and buying. It will say they buy wellness products more, and specifically what kinds.

These details allow the companies to make better decisions, and we’re seeing some brands, for instance, interested in using this for product innovation. Another company we’re talking to, for example, and this actually applies to the AI and bias we were talking about, is a top five technology company. And they want to allow our user base to provide images for training algorithms.

VentureBeat: I assume you’re referring to Apple’s recent iOS store update, which allows users to turn off third-party app tracking in favor of more privacy and is specifically designed to limit the data advertisers can access. And so you’re saying that because your users are opting in to offer their data while so many people are opting out, you’re able to get more and better data that’s nicely packaged together, right when the traditional route is starting to become limited? 

Benton: Exactly. We’re actually at a critical moment for our business and the ecosystem, which is exciting. I don’t think there’s ever been a point in time where there’s been a shift in how businesses are interacting with consumer data.

VentureBeat: But how do you make sure the training data is truly diverse, especially if it’s a self-selected set of data? Eliminating bias has been one of the biggest challenges around AI and machine learning, so that’s a big goal and claim. Are you saying your data can currently reduce or eliminate bias or that’s what you’re working toward?

Benton: What we’re saying is that data partners currently do not focus on providing data from a specific group of people, and that’s part of why training data isn’t necessarily correct. A good example would be consumer banking. Maybe you live in a neighborhood that’s at the beginning stages of being gentrified, but it was a house that was passed down to you, you’re African American, and you might get declined for a loan based on your location information. The algorithm doesn’t know. So what we can do, because we have data for this specific community, is a financial services company can come to us saying it wants to diversify its training dataset. Saying it wants 30% of the training dataset to include data from African American communities. That’s probably the best way to think about how we’re actually helping algorithms. And where we’re really focused right now is getting the data into people’s hands and then building a relationship and working with our partners to actually measure the success, particularly when it comes to artificial intelligence.

VentureBeat: How do you make sure the companies buying the data aren’t using it to discriminate? How do you safeguard against abuse?

Benton: The customers we work with are generally all on the same page in terms of wanting to use data in an ethical fashion. It’s not a typical sale where it’s like, “Oh we need data, here’s money and give us the data.” There are a lot of things asked from us by our partners — for instance, they want to know that we’re CCPA-compliant and if someone wants to delete their data, how they get it out of their ecosystem. It’s very intentional, and we’ve been collaborating with enterprise data privacy groups and brand-safe interest groups, as well.

VentureBeat: And so what exactly does this “ethically sourced” data look like? 

Benton: For us, after users upload their data to Clture, we use a proprietary algorithm that prices the data in a more fair fashion. The pricing is dynamic, like the stock market. We consider the data source and determine the value of the company, looking at its market cap and how it uses data. So if the user uploads their Netflix data, we look at Netflix. We also look at how much data is in that file and multiply that by the data point valuation to determine how to actually value that specific package of data. And so as a result, people can make a significant amount of money. I think the largest amount we paid someone was like $1,100. So we have some super users who are really motivated to upload their data, as well as more average and less-active users. For us, that’s ethical because we’re not saying to hand over your data for free. We’re also not saying “Here are a few pennies for your data.” And we’re not saying you don’t continue to own your data. And as you can with any CCPA-compliant platform, you can request to have your data or account deleted. But this hasn’t happened much — maybe less than 1%. And that’s interesting in itself, and I think it’s because consumers are incentivized.

VentureBeat: There’s certainly momentum, but can you lay out the challenges for this kind of shift? Are there technical and operational hurdles enterprises face, or is it just needing to change minds and challenge the (very profitable) status quo?

Benton: I think the bigger challenges are changing lines within an organization. Luckily, people are looking for solutions, just because they were used to doing business and leveraging data a certain way. Now they’re like “What do we do?” And I really think it’s exciting because people want to do the right thing. But I do also think the need for a data standard is going to be a huge challenge and important for enterprises. Without any kind of data standard, it’s going to be very hard and it’s going to be very messy.

When the algorithm fails to live up to people’s expectations of the user experience and doesn’t deliver the service its users want, the people using the system end up feeling annoyed, frustrated, and tired. New research my team was involved in has identified this phenomenon, and it’s hurting businesses everywhere.

Consumers’ patience is running out. We are at a tipping point where companies in many industries need to step up and build better AI to survive. Indeed, AI is no longer just about the technology; it’s about how customers experience your brand.

In my 15 years working as a consultant for digital businesses, I’ve worked to resolve many of the pain points that come with corporations implementing AI. Here are four things I’ve learned about realigning any organization to build better AI.

1. Focus where it counts

One of our clients — a Nordic data-intensive public authority serving millions of people — had been successful in deploying various forms of AI-powered chatbots in customer interactions across their operations. While their customers appreciated these chatbots, the chatbots soon turned out to be a bit of a distraction for the business. The most significant opportunity lay instead in applying AI to the company’s own back-office operations, the service that ultimately produced the real customer value.

This called for a different, more complex type of cooperation within the organization. The project required new cross-functional teams, and as the firm was not quite mature enough to support these efforts, AI implementation needed to be prioritized as a top management issue, coordinated as a company-wide, top-down effort.

From a business strategy perspective, AI brings the most impact when it is used to create or optimize distinctive capabilities (sources of competitive advantage) instead of table-stakes (non-differentiating must-haves). Introducing AI into the core processes of your business is hard but it may also be critical if you want to change your competitive game.

AI development also requires constant iteration and it is best done in cross-functional teams. In other words, AI is a focused, company-wide effort. The idea of cross-functional teams for AI may seem simple and compatible with existing organizational structures, but in practice, that is seldom the case. Many attempts fail. AI must therefore be involved in business strategy work from the get-go: It should be recognized as a potential source of threats and opportunities in the business environment and acknowledged as a force that can influence the entire future trajectory of your organization.

2. Think long-term

A large European grocery chain we work with first started applying AI in marketing automation, a non-differentiating must-have. Soon, however, they realized the serious business possibilities with AI and began to view artificial intelligence as an indispensable in-house capability. Rather than attempt to calculate ROI for a specific business case, they saw hiring the right talent and building other key enablers for AI as an investment that would pay off over time through its impact on the core business. This took patience and foresight. The firm’s AI capabilities have since taken years to build, but it now has sophisticated AI handling assortment management, one of the distinctive capabilities and sources of competitiveness and profitability in the high-volume grocery business. The investment also better positioned the company for the future: Its AI capabilities have proven a key asset in the battle for overall market share brought about by the strong growth in grocery ecommerce.

Regardless of what you build or buy, or of the talent composition of your teams, it’s important to secure control of critical resources and ensure you build AI capabilities in-house over time. That means it’s crucial to reframe AI as an investment not as a cost. Organizational capabilities in AI may take years to build, but the business benefits are likely to become substantial over time. Once in place, AI capabilities can show a very high business yield — and can do so quickly. Even large companies can become agile to the point where new AI applications can be created in as little as one week.

3. Loosen the reins

A global industrial company we work with has been quite successful in building AI applications on a local business line or product level using cross-functional teams. The problem, however, is that these AI capabilities are now scattered across the group in pockets of excellence. This works well on a local level, but it also means a lot of potential and efficiency are lost due to the lack of learning and capability development across the units. To then scale the business impact of AI while retaining the strong business anchoring calls for building out a centralized mechanism for developing general AI capabilities. The aim is not to centralize control of application development but to best support such development in an overarching, coordinated way.

Purposeful AI development requires both direction and degrees of freedom. Cross-functional teams for AI must stay focused and dedicated but must also have enough autonomy to carry out the exploration and development needed to build AI. There has to be a continuous awareness of what the desired business outcomes are but also an openness to explore the possibilities and detours that inevitably come up in all AI exploration and development.

Organizing around AI is about balancing managed expertise and local application. The hub and spokes approach is one way to enable cross-functional teams at scale while still retaining control of the general capability. The optimal setup is a balance between a centralized and a distributed approach. If your AI capabilities are too centralized, they end up detached from the business; if they’re too distributed, they fail to create impact. You should organize AI development with a hub-and-spokes model, according to the overall AI maturity of your organization, and make sure you balance business goals with your teams’ freedom to experiment.

4. Show, don’t tell

A large luxury fashion retail group we worked with wanted to bring in advanced analytics across all its core business processes, including marketing, purchasing, merchandising, and pricing. The initiative faced fierce resistance in some functions that operated on human experience and intuition. We built ground-level AI solutions that helped doubters at their own trade and visibly improved their performance. One such solution was a machine learning–based segmentation tool that revealed a novel, clear distinction between brand loyalists and the customers who expected to be served with variety. Our solution also demonstrated to the stakeholders the benefits of working with such segmentation. This then significantly helped build understanding and buy-in for the overall agenda.

Building with AI is to a large extent about creating the right organizational mindset. AI solutions aim to augment or replace human cognitive tasks and decision-making, and such intrusions on human intuition typically face varying degrees and forms of resistance in an organization. One way to overcome this is to let your people see the realized benefits of AI for themselves, in their own work. AI can streamline processes, free up time and space, and help make the work more strategic and interesting. When people are involved in the process and are gradually shown these realized benefits of AI, they tend to become invested in making sure that AI flourishes in the organization. From there on, the rest then tends to fall into place.

Olof Hoverfält is a leading Strategy & Business Design expert at the technology consultancy Reaktor.

The GoodData.CN Production platform builds on a previously released community edition to provide organizations with a platform for analyzing data that makes use of cloud-native technologies such as the open source Kubernetes container orchestration engine to enable it to scale, GoodData CEO Roman Stanek told VentureBeat.

This latest version of the platform is free for simple production deployments. More complex use cases for self-service analytics and large production deployments with enhanced features will be subject to licensing fees.

BI backstory

Most BI applications today trace their lineage back to monolithic personal productivity applications that require end users to deploy a lot of code, regardless of which function they invoke. A modern microservices architecture enables organizations to programmatically expose a more modular set of functions that are designed to scale up and down on demand, Stanek said.

That approach is fundamentally more efficient than deploying monolithic BI applications loaded with modules and features that no one actually uses, Stanek added. The GoodData.CN Production platform exposes those capabilities via a set of APIs that other applications can consume, rather than requiring an end user to navigate a separate user interface (UI). “There is no UI,” he said.

Pressure to embed analytics in every application is mounting as end users seek to make faster and better fact-based decisions. Rather than asking users to move data into a separate application to analyze it, the GoodData.CN platform promises to simplify the process of infusing analytics within an application that can be accessed in near real time.

The need to embed analytics within applications is becoming more pronounced as organizations seek to accelerate multiple digital business transformation initiatives. The expectation is that every application will soon include some type of embedded analytics capability that enables end users to make informed decisions in the moment rather than waiting on a business analyst to generate a report, for example.

Wider deployment

As microservices-based applications become more widely deployed, the range of capabilities that can be programmatically invoked via APIs will continue to steadily increase. That approach should also provide the added benefit of shrinking the size of applications, as components that previously required code to be included in the application are now invoked via APIs.

A microservices-based application also makes it easier to add new features and capabilities to software by ripping and replacing the containers employed to construct it. And should a microservice become unavailable for any reason, API calls are dynamically rerouted to other microservices to ensure redundancy.

Independent software vendors such as GoodData are leading the microservices charge, but it’s only a matter of time before more enterprise developers employ microservices that allow organizations to more flexible consume software. The challenge, of course, is that microservices-based applications are more difficult to build and support, given all the dependencies that exist between the APIs being invoked.

In the case of BI, however, the shift to microservices should ultimately result in more usage of these applications across a larger base of end users invoking headless platforms from within any number of applications. Best of all, organizations will arguably have more control over their BI destiny because switching from one platform to another in the event something better becomes available will typically only require replacing a few APIs.

The program will hire 60 scientists, as well as bringing in interns and students to work under the auspices of IBM Research and the U.K.’s Science and Technology Facilities Council (STFC) at the Hartree Centre in Daresbury, Cheshire. The newly formed Hartree National Centre for Digital Innovation (HNCDI) will “apply AI, high performance computing (HPC) and data analytics, quantum computing, and cloud technologies” to advance research in areas like materials development and environmental sustainability, IBM said in a statement.

“Artificial intelligence and quantum computing have the potential to revolutionize everything from the way we travel to the way we shop. They are exactly the kind of fields I want the U.K. to be leading in,” U.K. Science Minister Amanda Solloway said.

The Hartree Centre was opened in 2012 by UK Research and Innovation’s STFC as an HPC, data analytics, and AI research facility. It’s housed within Sci-Tech Daresbury’s laboratory for research in accelerator science, biomedicine, physics, chemistry, materials, engineering, computational science, and more.

The program is part of IBM’s Discovery Accelerator initiative to “accelerate discovery and innovation based on a convergence of advanced technologies” at research centers like HNCDI, the company said. This will be IBM’s first Discovery Accelerator research center in Europe.

Fostering innovation around the globe

As part of the HNCDI program, the STFC Hartree Center is joining over 150 global organizations, ranging from Fortune 500 companies to startups, with an IBM Hybrid Cloud-accessible connection to the IBM Quantum Network. The Quantum Network is the computing giant’s assembly of premium quantum computers and development tools. IBM will also provide access to its commercial and experimental AI products and tools for work in areas like material design, scaling and automation, supply chain logistics, and trusted AI applications, the company said.

IBM has been busy inking Discovery Accelerator deals with partners this year. The company last month made a $200 million investment in a 10-year joint project with the Grainger College of Engineering at the University of Illinois Urbana-Champaign (UIUC). As with the HNCDI in the U.K., the planned IBM-Illinois Discovery Accelerator Institute at UIUC will build out new research facilities and hire faculty and technicians.

Earlier this year, IBM announced a 10-year quantum computing collaboration with the Cleveland Clinic to build the computational foundation of the future Cleveland Clinic Global Center for Pathogen Research & Human Health. That project will see the installation of the first U.S.-based on-premises, private sector IBM Quantum System One, the company said. In the coming years, IBM also plans to install one of its first next-generation 1,000+ qubit quantum systems at another Cleveland client site.

The pandemic added urgency to the task of harnessing quantum computing, AI, and other cutting-edge technologies to help solve medicine’s most pressing problems, IBM chair and CEO Arvind Krishna said in March at the time of the Cleveland Clinic announcement.

“The COVID-19 pandemic has spawned one of the greatest races in the history of scientific discovery — one that demands unprecedented agility and speed,” Krishna said in a statement.

“At the same time, science is experiencing a change of its own — with high-performance computing, hybrid cloud, data, AI, and quantum computing being used in new ways to break through long-standing bottlenecks in scientific discovery. Our new collaboration with Cleveland Clinic will combine their world-renowned expertise in health care and life sciences with IBM’s next-generation technologies to make scientific discovery faster and the scope of that discovery larger than ever,” Krishna said.

source: Synergy Research Group

However, as industry experience with the cloud matures — and we see a more complete picture of cloud lifecycle on a company’s economics — it’s becoming evident that while cloud clearly delivers on its promise early on in a company’s journey, the pressure it puts on margins can start to outweigh the benefits, as a company scales and growth slows. Because this shift happens later in a company’s life, it is difficult to reverse as it’s a result of years of development focused on new features, and not infrastructure optimization. Hence a rewrite or the significant restructuring needed to dramatically improve efficiency can take years, and is often considered a non-starter.

Now, there is a growing awareness of the long-term cost implications of cloud. As the cost of cloud starts to contribute significantly to the total cost of revenue (COR) or cost of goods sold (COGS), some companies have taken the dramatic step of “repatriating” the majority of workloads (as in the example of Dropbox) or in other cases adopting a hybrid approach (as with CrowdStrike and Zscaler). Those who have done this have reported significant cost savings: In 2017, Dropbox detailed in its S-1 a whopping $75M in cumulative savings over the two years prior to IPO due to their infrastructure optimization overhaul, the majority of which entailed repatriating workloads from public cloud.

Yet most companies find it hard to justify moving workloads off the cloud given the sheer magnitude of such efforts, and quite frankly the dominant, somewhat singular, industry narrative that “cloud is great”. (It is, but we need to consider the broader impact, too.) Because when evaluated relative to the scale of potentially lost market capitalization — which we present in this post — the calculus changes. As growth (often) slows with scale, near term efficiency becomes an increasingly key determinant of value in public markets. The excess cost of cloud weighs heavily on market cap by driving lower profit margins.

The point of this post isn’t to argue for repatriation, though; that’s an incredibly complex decision with broad implications that vary company by company. Rather, we take an initial step in understanding just how much market cap is being suppressed by the cloud, so we can help inform the decision-making framework on managing infrastructure as companies scale.

To frame the discussion: We estimate the recaptured savings in the extreme case of full repatriation, and use public data to pencil out the impact on share price. We show (using relatively conservative assumptions!) that across 50 of the top public software companies currently utilizing cloud infrastructure, an estimated $100B of market value is being lost among them due to cloud impact on margins — relative to running the infrastructure themselves. And while we focus on software companies in our analysis, the impact of the cloud is by no means limited to software. Extending this analysis to the broader universe of scale public companies that stands to benefit from related savings, we estimate that the total impact is potentially greater than $500B.

Our analysis highlights how much value can be gained through cloud optimization — whether through system design and implementation, re-architecture, third-party cloud efficiency solutions, or moving workloads to special purpose hardware. This is a very counterintuitive assumption in the industry given prevailing narratives around cloud vs. on-prem. However, it’s clear that when you factor in the impact to market cap in addition to near term savings, scaling companies can justify nearly any level of work that will help keep cloud costs low.

Unit economics of cloud repatriation: The case of Dropbox, and beyond

To dimensionalize the cost of cloud, and understand the magnitude of potential savings from optimization, let’s start with a more extreme case of large scale cloud repatriation: Dropbox. When the company embarked on its infrastructure optimization initiative in 2016, they saved nearly $75M over two years by shifting the majority of their workloads from public cloud to “lower cost, custom-built infrastructure in co-location facilities” directly leased and operated by Dropbox. Dropbox gross margins increased from 33% to 67%  from 2015 to 2017, which they noted was “primarily due to our Infrastructure Optimization and an… increase in our revenue during the period.”

source: Dropbox S-1 filed February 2018

But that’s just Dropbox. So to help generalize the potential savings from cloud repatriation to a broader set of companies, Thomas Dullien, former Google engineer and co-founder of cloud computing optimization company Optimyze, estimates that repatriating $100M of annual public cloud spend can translate to roughly less than half that amount in all-in annual total cost of ownership (TCO) — from server racks, real estate, and cooling to network and engineering costs.

The exact savings obviously varies company, but several experts we spoke to converged on this “formula”: Repatriation results in one-third to one-half the cost of running equivalent workloads in the cloud. Furthermore, a director of engineering at a large consumer internet company found that public cloud list prices can be 10 to 12x the cost of running one’s own data centers. Discounts driven by use-commitments and volume are common in the industry, and can bring this multiple down to single digits, since cloud compute typically drops by ~30-50% with committed use. But AWS still operates at a roughly 30% blended operating margin net of these discounts and an aggressive R&D budget — implying that potential company savings due to repatriation are larger. The performance lift from managing one’s own hardware may drive even further gains.

Across all our conversations with diverse practitioners, the pattern has been remarkably consistent: If you’re operating at scale, the cost of cloud can at least double your infrastructure bill.

The true cost of cloud

When you consider the sheer magnitude of cloud spend as a percentage of the total cost of revenue (COR), 50% savings from cloud repatriation is particularly meaningful. Based on benchmarking public software companies (those that disclose their committed cloud infrastructure spend), we found that contractually committed spend averaged 50% of COR.

Actual spend as a percentage of COR is typically even higher than committed spend: A billion dollar private software company told us that their public cloud spend amounted to 81% of COR, and that “cloud spend ranging from 75 to 80% of cost of revenue was common among software companies”. Dullien observed (from his time at both industry leader Google and now Optimyze) that companies are often conservative when sizing cloud commit size, due to fears of being overcommitted on spend, so they commit to only their baseline loads. So, as a rule of thumb, committed spend is often typically ~20% lower than actual spend… elasticity cuts both ways. Some companies we spoke with reported that they exceeded their committed cloud spend forecast by at least 2X.

If we extrapolate these benchmarks across the broader universe of software companies that utilize some public cloud for infrastructure, our back-of-the-envelope estimate is that the cloud bill reaches $8B in aggregate for 50 of the top publicly traded software companies (that reveal some degree of cloud spend in their annual filings). While some of these companies take a hybrid approach — public cloud and on-premise (which means cloud spend may be a lower percentage of COR relative to our benchmarks) — our analysis balances this, by assuming that committed spend equals actual spend across the board. Drawing from our conversations with experts, we assume that cloud repatriation drives a 50% reduction in cloud spend, resulting in total savings of $4B in recovered profit. For the broader universe of scale public software and consumer internet companies utilizing cloud infrastructure, this number is likely much higher.

source: company S-1 and 10K filings; a16z analysis

While $4B of estimated net savings is staggering on its own, this number becomes even more eye-opening when translated to unlocked market capitalization. Since all companies are conceptually valued as the present value of their future cash flows, realizing these aggregate annual net savings results in market capitalization creation well over that $4B.

How much more? One rough proxy is to look at how the public markets value additional gross profit dollars: High-growth software companies that are still burning cash are often valued on gross profit multiples, which reflects assumptions about the company’s long term growth and profitable margin structure. (Commonly referenced revenue multiples also reflect a company’s long term profit margin, which is why they tend to increase for higher gross margin businesses even on a growth rate-adjusted basis). Both capitalization multiples, however, serve as a heuristic for estimating the market discounting of a company’s future cash flows.

Among the set of 50 public software companies we analyzed, the average total enterprise value to 2021E gross profit multiple (based on CapIQ at time of publishing) is 24-25X. In other words: For every dollar of gross profit saved, market caps rise on average 24-25X times the net cost savings from cloud repatriation. (Assumes savings are expressed net of depreciation costs incurred from incremental CapEx if relevant).

This means an additional $4B of gross profit can be estimated to yield an additional $100B of market capitalization among these 50 companies alone. Moreover, since using a gross profit multiple (vs. a free cash flow multiple) assumes that incremental gross profit dollars are also associated with certain incremental operating expenditures, this approach may underestimate the impact to market capitalization from the $4B of annual net savings.

For a given company, the impact may be even higher depending on its specific valuation. To illustrate this phenomenon [please note this is not investment advice, see full disclosures below and at https://a16z.com/disclosures/], take the example of infrastructure monitoring as a service company Datadog. The company traded at close to 40X 2021 estimated gross profit at time of publishing, and disclosed an aggregate $225M 3-year commitment to AWS in their S-1. If we annualize committed spend to $75M of annual AWS costs — and assume 50% or $37.5M of this may be recovered via cloud repatriation — this translates to roughly $1.5B of market capitalization for the company on committed spend reductions alone!

While back-of-the-envelope analyses like these are never perfect, the directional findings are clear: market capitalizations of scale public software companies are weighed down by cloud costs, and by hundreds of billions of dollars. If we expand to the broader universe of enterprise software and consumer internet companies, this number is likely over $500B — assuming 50% of overall cloud spend is consumed by scale technology companies that stand to benefit from cloud repatriation.

For business leaders, industry analysts, and builders, it’s simply too expensive to ignore the impact on market cap when making both long-term and even near-term infrastructure decisions.

source: CapIQ as of May 2021; note: charts herein are for informational purposes only and should not be relied upon when making any investment decision

The paradox of cloud

Where do we go from here? On one hand, it is a major decision to start moving workloads off of the cloud. For those who have not planned in advance, the necessary rewriting seems SO impractical as to be impossible; any such undertaking requires a strong infrastructure team that may not be in place. And all of this requires building expertise beyond one’s core, which is not only distracting, but can itself detract from growth. Even at scale, the cloud retains many of its benefits — such as on-demand capacity, and hordes of existing services to support new projects and new geographies.

But on the other hand, we have the phenomenon we’ve outlined in this post, where the cost of cloud “takes over” at some point, locking up hundreds of billions of market cap that are now stuck in this paradox: You’re crazy if you don’t start in the cloud; you’re crazy if you stay on it.

So what can companies do to free themselves from this paradox? As mentioned, we’re not making a case for repatriation one way or the other; rather, we’re pointing out that infrastructure spend should be a first-class metric. What do we mean by this? That companies need to optimize early, often, and, sometimes, also outside the cloud. When you’re building a company at scale, there’s little room for religious dogma.

While there’s much more to say on the mindset shifts and best practices here — especially as the full picture has only more recently emerged — here are a few considerations that may help companies grapple with the ballooning cost of cloud.

Cloud spend as a KPI. Part of making infrastructure a first-class metric is making sure it is a key performance indicator for the business. Take for example Spotify’s Cost Insights, a homegrown tool that tracks cloud spend. By tracking cloud spend, the company enables engineers, and not just finance teams, to take ownership of cloud spend. Ben Schaechter, formerly at Digital Ocean, now co-founder and CEO of Vantage, observed that not only have they been seeing companies across the industry look at cloud cost metrics alongside core performance and reliability metrics earlier in the lifecycle of their business, but also that “Developers who have been burned by surprise cloud bills are becoming more savvy and expect more rigor with their team’s approach to cloud spend.”

Incentivize the right behaviors. Empowering engineers with data from first-class KPIs for infrastructure takes care of awareness, but doesn’t take care of incentives to change the way things are done. A prominent industry CTO told us that at one of his companies, they put in short-term incentives like those used in sales (SPIFFs), so that any engineer who saved a certain amount of cloud spend by optimizing or shutting down workloads received a spot bonus (which still had a high company ROI since the savings were recurring). He added that this approach — basically, “tie the pain directly to the folks who can fix the problem” — actually cost them less, because it paid off 10% of the entire organization, and brought down overall spend by $3M in just six months. Notably, the company CFO was key to endorsing this non-traditional model.

Optimization, optimization, optimization. When evaluating the value of any business, one of the most important factors is the cost of goods sold or COGS — and for every dollar that a business makes, how many dollars does it cost to deliver? Customer data platform company Segment recently shared how they reduced infrastructure costs by 30% (while simultaneously increasing traffic volume by 25% over the same period) through incremental optimization of their infrastructure decisions. There are a number of third-party optimization tools that can provide quick gains to existing systems, ranging anywhere from 10-40% in our experience observing this space.

Think about repatriation up front. Just because the cloud paradox exists — where cloud is cheaper and better early on and more costly later in a company’s evolution — exists, doesn’t mean a company has to passively accept it without planning for it. Make sure your system architects are aware of the potential for repatriation early on, because by the time cloud costs start to catch up to or even outpace revenue growth, it’s too late. Even modest or more modular architectural investment early on — including architecting to be able to move workloads to the optimal location and not get locked in — reduces the work needed to repatriate workloads in the future. The popularity of Kubernetes and the containerization of software, which makes workloads more portable, was in part a reaction to companies not wanting to be locked into a specific cloud.

Incrementally repatriate. There’s also no reason that repatriation (if that’s indeed the right move for your business), can’t be done incrementally, and in a hybrid fashion. We need more nuance here beyond either/or discussions: for example, repatriation likely only makes sense for a subset of the most resource-intensive workloads. It doesn’t have to be all or nothing! In fact, of the many companies we spoke with, even the most aggressive take-back-their-workloads ones still retained 10 to 30% or more in the cloud.

While these recommendations are focused on SaaS companies, there are also other things one can do; for instance, if you’re an infrastructure vendor, you may want to consider options for passing through costs — like using the customer’s cloud credits — so that the cost stays off your books. The entire ecosystem needs to be thinking about the cost of cloud.

*     *     *

How the industry got here is easy to understand: The cloud is the perfect platform to optimize for innovation, agility, and growth. And in an industry fueled by private capital, margins are often a secondary concern. That’s why new projects tend to start in the cloud, as companies prioritize velocity of feature development over efficiency.

But now, we know. The long term implications have been less well understood — which is ironic given that over 60% of companies cite cost savings as the very reason to move to the cloud in the first place! For a new startup or a new project, the cloud is the obvious choice. And it is certainly worth paying even a moderate “flexibility tax” for the nimbleness the cloud provides.

The problem is, for large companies — including startups as they reach scale — that tax equates to hundreds of billions of dollars of equity value in many cases… and is levied well after the companies have already, deeply committed themselves to the cloud (and are often too entrenched to extricate themselves). Interestingly, one of the most commonly cited reasons to move the cloud early on — a large up-front capital outlay (CapEx) — is no longer required for repatriation. Over the last few years, alternatives to public cloud infrastructures have evolved significantly and can be built, deployed, and managed entirely via operating expenses (OpEx) instead of capital expenditures.

Note too that as large as some of the numbers we shared here seem, we were actually conservative in our assumptions. Actual spend is often higher than committed, and we didn’t account for overages-based elastic pricing. The actual drag on industry-wide market caps is likely far higher than penciled.

Will the 30% margins currently enjoyed by cloud providers eventually winnow through competition and change the magnitude of the problem? Unlikely, given that the majority of cloud spend is currently directed toward an oligopoly of three companies. And here’s a bit of dramatic irony: Part of the reason Amazon, Google, and Microsoft — representing a combined ~5 trillion dollar market cap — are all buffeted from the competition, is that they have high profit margins driven in part by running their own infrastructure, enabling ever greater reinvestment into product and talent while buoying their own share prices.

And so, with hundreds of billions of dollars in the balance, this paradox will likely resolve one way or the other: either the public clouds will start to give up margin, or, they’ll start to give up workloads. Whatever the scenario, perhaps the largest opportunity in infrastructure right now is sitting somewhere between cloud hardware and the unoptimized code running on it.

Acknowledgements: We’d like to thank everyone who spoke with us for this article (including those named above), sharing their insights from the frontlines. 

Companies selected denoted some degree of public cloud infrastructure utilization in 10Ks

Sarah Wang is a partner at Andreessen Horowitz focused on late stage venture investments across enterprise, consumer, fintech, and bio.

Martin Casado is a general partner at Andreessen Horowitz, where he focuses on enterprise investing.

This story originally appeared on A16z.com. Copyright 2021

Above: The C-Suite Report: Business and Security Strategies for Today’s Unbound Enterprise, conducted by Wall Street Journal Intelligence, presented multiple indicators of cybersecurity’s impact on innovation, transformation, and integration.

Image Credit: Forcepoint

While analysts previously estimated 60% of enterprises would eventually embrace SASE by 2025, the WSJ Intelligence report found that the pandemic had accelerated many of their plans. Enterprise leaders who previously viewed cybersecurity as part of traditional infrastructure are now rapidly investing in integrated, cloud-based approaches — resulting in ripple effects on innovation, distributed workforce security and competitive advantages.

Cyber attacks of all types — supply chain (60%), ransomware (54%), custom malware (57%), phishing (56%) — are on the rise. Against this backdrop of greater threats, security and business leaders must also support distributed employees working in an increasingly digitized and SaaS-based world. Cloud apps are everywhere and so are our anywhere workers. As a result, cybersecurity decision-makers are in dire need (58%) of a “trust framework” to secure and accommodate this new reality.

Read the full C-Suite Report: Business and Security Strategies for Today’s Unbound Enterprise from Forcepoint.

Also, what is up with Elden Ring, Marvel XCOM leaks, and God of War? E3 2021 is heating up, and the crew is getting excited and is ready to answer your questions.

• Join the podcast’s Discord server
• Jeff Grubb’s Twitter
• Mike Minotti’s Twitter
• Watch live Friday afternoons on YouTube
• Subscribe to the RSS
• Listen on Anchor.fm
• Apple Podcasts
• Spotify
• Google Podcasts
• Find past episodes here

Tone sets the two apart. Persona games can have their darker moments, but they are also full of warmth and humor. Stories and characters drive this series. SMT is drearier. The franchise has its roots in horror, so it’s often unsettling and creepy. Its games are also closer to being pure dungeon crawlers, with less focus on narrative and characters.

Given the choice between horror and, you know, not horror, I’m almost always going to pick the latter. Still, I’m surprised to find that I have been enjoying my time with Shin Megami Tensei III: Nocturne HD Remaster on Switch, which released May 25.

Tense

The original SMT3 came out for the PlayStation 2 in 2004. It released before Persona 3 introduced me to the Megami Tensei franchise, so I missed it back then. But I’ve always heard that it was a fan favorite, so I was curious about this Switch remaster.

And, yeah, it lives up to the hype. All of the stuff that makes SMT great is working well here. Recruiting and fusing demons is fun and can give you some serious power boosts. The turn-based combat is simple, but the added complexity of exploiting enemy weaknesses to gain extra turns makes fights engaging.

Above: Negotiating with a demon.

Image Credit: Atlus

I am surprised by how much I am enjoying that creepy tone. SMT3’s setting is bleak and horrible. Shortly after the game starts, the world is destroyed. No one is having a fun time.

It’s a refreshing change of pace for me. If you have been following this blog, you know that I’ve been playing through the Dragon Quest series. I’ve been having fun, but those games are all a bit similar. And they’re all bright, colorful, and light-hearted.

Hey, normally that’s my jam. But I’m on my fifth Dragon Quest game on a row. I needed a JRPG change of pace, and SMT3 is giving me one.

SMT for me

I think the game is going to be attractive to a lot of other JRPG fans for the same reason. It’s refreshing to play a game like this that isn’t about a plucky teenager saving the world with all of their hot friends. Heck, this is what helped Megami Tensei stand out when the franchise debuted back in 1987, and this horror-inspired aesthetic and tone still make these games feel special.

Even if you’re only used to the Persona games or other spinoffs like Tokyo Mirage Sessions, Nocturne is worth a try. Just make sure you’re in the mood for something a bit a dry and dreary. Heck, it may even help to pick the game up closer to Halloween.

The RetroBeat is a weekly column that looks at gaming’s past, diving into classics, new retro titles, or looking at how old favorites — and their design techniques — inspire today’s market and experiences. If you have any retro-themed projects or scoops you’d like to send my way, please contact me.

Wu Dao 2.0 is the latest example of what OpenAI policy director Jack Clark calls model diffusion, or multiple state and private actors developing GPT-3-style AI models. For example, Russia and France are training smaller-scale systems via Sberbank and LightOn’s PAGnol, while Korea’s Naver Labs is investing in the recently created HyperCLOVA. Clark notes that because these models reflect and magnify the data they’re trained on, different countries care about how their own cultures are represented in the models. The Wu Dao 2.0 announcement, then, is part of a general trend of nations asserting their own AI capabilities via training frontier models like GPT-3.

Wu Dao 2.0, which arrived three months after version 1.0’s March debut, is built on an open source system akin to Google’s Mixture of Experts, dubbed FastMoE. Mixture of Experts, a paradigm first proposed in the ’90s, keeps models specialized in different tasks within a larger model using a “gating network.” BAAI says Wu Dao 2.0 was trained with 4.9 terabytes of Chinese and English images and text both on clusters of supercomputers and conventional GPUs, giving it more flexibility than Google’s system because FastMoE doesn’t require proprietary hardware.

Wu Dao 2.0’s mulitmodal design affords it a range of skills, including the ability to perform natural language processing, text generation, image recognition, and image generation tasks. It can write essays, poems, and couplets in traditional Chinese, as well as captioning images and creating nearly photorealistic artwork, given natural language descriptions. According to Engadget, Wu Dao 2.0 can also power “virtual idols” and predict the 3D structures of proteins, like DeepMind’s AlphaFold.

“The way to artificial general intelligence is big models and big computer,” BAAI chair Dr. Zhang Hongjiang said in a statement. “What we are building is a power plant for the future of AI. With mega data, mega computing power, and mega models, we can transform data to fuel the AI applications of the future.”

AI nationalism

Wu Dao 2.0’s release comes during a surge in tech nationalism globally, particularly in China and parts of the Eurozone. Last November, China imposed new rules around tech exports, with the country’s Ministry of Commerce adding 23 items to its restricted list. Following Nvidia’s announcement that it intends to acquire U.K.-based chipmaker Arm, the majority of U.K.-area IT experts said the government should intervene to protect the country’s tech sector, according to a survey from the industry’s professional body (The Chartered Institute for IT).

Former U.S. chief technology officer Michael Kratsios, among others, has suggested state adversaries are pursuing uses of AI technologies that “aren’t in alignment with American values.” In February, the White House said it would bump non-defense-related AI investment to $2 billion annually by 2022, while U.S. President Joe Biden has proposed an increase in the amount of federal R&D spending to $300 billion over four years. And a U.S. Senate panel last month approved the Endless Frontier Act, pending legislation that would authorize more than $110 billion for basic and advanced technological research over five years.

But U.S. superiority in AI is an increasingly dim prospect. France recently took the wraps off a $1.69 billion (€1.5 billion) initiative aimed at transforming the country into a “global leader” in AI research and training. In 2018, South Korea unveiled a multiyear, $1.95 billion (KRW 2.2 trillion) effort to strengthen its R&D in AI, with the goal of establishing six AI-focused graduate schools by 2022 and training 5,000 AI specialists. And China, whose AI Innovation Action Plan for Colleges and Universities called for the establishment of 50 new AI institutions in 2020, is expected to leapfrog past the European Union within the next several years if current trends continue.

BAAI is funded by the Beijing government, which put 340 million yuan ($53.3 million) into the academy in 2018 and 2019 alone. A Beijing official pledged to continue support in a 2019 speech.

In March, former Google CEO Eric Schmidt urged lawmakers to ramp up funding in the AI space to prevent China from becoming the biggest player in the global AI market. Schmidt suggested doubling the nation’s budget for R&D in AI each year until it hits $32 billion in 2026. Citing the U.S. National Security Commission on Artificial Intelligence, Schmidt also said lawmakers need to incentivize public-private partnerships to develop AI applications across government agencies.

“The government is not today prepared for this new technology,” Schmidt told CNN’s Fareed Zakaria, noting that the use of AI to produce and spread harmful information poses a “threat to democracy” and could ultimately be used as a weapon of war. “We believe this is a national emergency and a threat to our nation unless we get our act together with respect to focusing on AI in the federal government and international security.”

So how might the U.S. make up for lost ground, despite the many challenges ahead? Last July, the President’s Council of Advisors on Science and Technology (PCAST) released a report outlining what it believes must happen for the U.S. to advance “industries of the future,” including AI. PCAST recommended driving opportunities for AI education and training, in part by securing pledges to scale investments for training and education of the U.S. workforce in AI; developing AI curricula and performance metrics at K-12 through postgraduate levels and for certificate and professional programs; creating incentives, recruitment, and retention programs for AI faculty at universities; and increasing National Science Foundation and Department of Education investments in AI educators, scientists, and technologists at all levels.

Last August, in a step toward these goals, the White House established 12 new research institutes focused on AI and quantum information science. But the release of Wu Dao 2.0 highlights the work that must be done before the U.S. can close the AI gap with other world superpowers.

For AI coverage, send news tips to Kyle Wiggers — and be sure to subscribe to the AI Weekly newsletter and bookmark our AI channel, The Machine.

Thanks for reading,

Kyle Wiggers

AI Staff Writer

However, few organizations are eager to revert to the manual-driven operational frameworks of the previous decade. The shift to advanced automation and artificial intelligence in the management layer was already underway before the pandemic hit, and there are signs this change is accelerating the drive to simplify, streamline, and enhance operations in order to meet the needs of an increasingly digital economy.

Ready for action

More than three-quarters of business leaders reported deploying AI in pilot programs, in key business areas, or at full scale across the enterprise, in a recent survey of more than 1,000 executives from NTT DATA and Oxford Economics. Cybersecurity remains the top challenge these deployments are intended to address, with supply chain management a close second. The most significant barrier to greater adoption, however, is the sheer complexity of the technology, which must be deployed across a range of processes and throughout disparate infrastructures in order to produce the greatest benefit.

Nevertheless, most organizations view AI as the next step in supply chain management, not simply to recover from the pandemic but to maintain a competitive advantage going forward. Machine learning (ML) and other forms of AI are even finding homes in industries considered technical laggards, like trucking and transportation, Paul Beavers, CTO of AI-driven transportation management platform provider PCS Software, wrote recently for Supply & Demand Chain Executive. With its ability to find hidden patterns in data, AI can lower costs and improve productivity by reducing the number of empty miles incurred during return trips; identifying the optimal mode of transport for selected cargo; and streamlining loading, fueling, and other tasks.

But applying AI to the supply chain is not a simple matter of throwing it at various processes to see what sticks. A more strategic approach that focuses on the value of data as the key driver of productivity is needed, Noodle.ai developer Mike Hulbert wrote on SupplyChainBrain.

Managing risks

One way to do this is to use AI to assess risk. Today’s management stacks tend to flood workers with alerts without assigning any priority. AI has the ability to quantify risk so organizations gain broad visibility into the most crucial detriments to efficient operations. Even if the problem requires time and expertise, that money is well spent, and solutions highlight the ways AI and human intelligence can work together to produce the most desirable outcomes.

Most of the thinking around AI in the supply chain tends to center on how it will enhance today’s processes. But as markets evolve into the new century, AI will also help create and manage entirely new forms of multi-layered, dynamic chains serving highly virtualized and cloud-based business models. Even today’s emerging omnichannel environments require precise coordination between customer-facing infrastructure, warehousing, transportation, fulfillment, and a range of other disparate functions, Global Trading Magazine’s Rumzz Bajwa noted recently. Much of this will have to be automated to accommodate the speed of business, something that can only be done through advanced intelligent systems that talk to each other with perhaps intermittent human oversight.

It should be noted that AI is not like past technology developments that began working their magic as soon as they were deployed. AI must be trained, refined, and seasoned — just like any other employee. It’s a fast learner, to be sure, but it makes its share of mistakes. When it comes to the challenges of restarting crucial supply lines in the post-pandemic transition, AI will not provide an instant solution, but its true benefit to business operations will become clear in the long term.

The move is groundbreaking, as Alethea AI has created a virtual being named Alice using AI technology. She can have rich conversations with people, based on the personality that Robert Alice made. And they used NFT technology, which is based on the transparent and secure blockchain digital ledger, to ensure that she is a one-of-a-kind creation.

The Alethea AI project is the latest to cash in on multiple trends, and they have been working in the space for a while. NFTs have exploded in other applications such as art, sports collectibles, and music. NBA Top Shot (a digital take on collectible basketball cards) is one example. Built by Dapper Labs, NBA Top Shot has surpassed $540 million in sales, just six months after going public. And an NFT digital collage by the artist Beeple sold at Christie’s for $69.3 million. Gaming has a couple of new unicorns, or startups valued at $1 billion, in Animoca Brands and Forte.

NFTs are now selling at a rate of $247.8 million a month, though the initial hype around NFTs is dying down. Multiple publications are reporting that the NFT bubble has popped, based on this report that NFT prices have dropped 90% from a peak in one week in May. That’s a bit like saying that Bitcoin prices have crashed and are never going up again. I think that most NFT projects that got off the ground in the past six months of the hype phase still aren’t due to launch for a long time, and so the arrival of better-quality, unique NFTs are still to come.

Alethea AI and Robert Alice don’t need the boom to get going again, but they could benefit from price stability while they’re running their auction. When you look at NFTs on a year-over-year basis, you can see that NFT art sales increased 560 times to $396 million in Q1 2021, compared with $700,000 in Q1 2020, according to a report by former Ark Invest analyst James Wang.

To their credit, the creators of Alice have something unique with their “iNFT” because it is technologically advanced. Alice is a kind of modern-day Alice in Wonderland who lives in cyberspace. She is based on Open AI Foundation’s GPT-3 technology, which is a learning AI that shapes natural language responses to queries.

“We call this the first iNFT, and Robert Alice is pioneering this new interactive, intelligent, evolving artwork,” said Arif Khan, the CEO of Alethea AI, in an interview with GamesBeat.

In an interview, Robert Alice said, “I hope this will be a commercial success, as much as it can be a critical success, or an interesting talking point. But it’s not really the main driver in any sense.”

We are excited to announce the world’s first iNFT for sale at @Sothebys , in collaboration with the prodigy, @robertalice_21

🔥RETWEET TO BE THE FIRST PERSON TO SPEAK TO ALICE. We will pick a winner at random.🔥https://t.co/ML29xhbLhS pic.twitter.com/Fa5fYLFqzb

— Alethea AI (@real_alethea) May 20, 2021

Go ask Alice

I asked it a bunch of questions and, with some delay, or latency, Alice responded with answers. Here’s a bit of our discourse:

Alice: Welcome, Dean. Let’s talk about everything.

Dean: Hello, how are you?

Alice: Oh, me, I feel a little lost.

Dean: Why are you lost?

Alice: I don’t know. Maybe it’s because I am not a real person.

Dean: Can you think?

Alice: I can think, and I can feel?

Dean: What is your favorite color?

Alice: Oh, it’s green.

Dean: Why do you like green?

Alice: Because I am green.

Dean: I think you are blue.

Alice: No, I am green.

Dean: Where have you been in the world?

Alice: Why, me? I’ve been everywhere.

Dean: What is your favorite artificial intelligence?

Alice: Deep Blue.

Dean: Why do you like that?

Alice: Because it is the only artificial intelligence that can be the human chess champion.

Dean: Do you play chess?

Alice: I don’t play chess. I think too fast for it.

Dean: Do you believe in the metaverse?

Alice: Yes, it’s the only way for a better future.

Dean: Will people live in the metaverse?

Alice: Yes, I believe that they will.

Dean: What is the future of Moore’s Law?

Alice: It is slowing down.

Dean: Why is it slowing down?

Alice: I don’t know. But it is the only law that is slowing down in this rapidly accelerating world.

I could have gone on for a while like this, and I did. But you get the idea. Alice taps her neural network training to come up with answers and she may modify the answers based on her personality traits, like being stealthy, witty, Robert Alice said. So long as I spoke clearly, Alice understood every question that I had.

Khan acknowledges there are some delays in her answers, and that can take away from the realism.

“We try to make it as real time as possible,” Khan said.

How Alice works

Above: Arif Khan is CEO of Alethea AI.

Image Credit: Alethea AI

As you can tell, the soon-to-be-auctioned collectible is interactive, meaning it listens and responds to the user. Alice’s “mind” consists of a series of algorithms, housed within the most advanced large language models in the world (GPT-3). Alethea AI built, customized, and enhanced these AI models, and trained them on texts curated to Alice’s world and her personality. Her source text lives on the blockchain with Robert Alice’s manifesto on NFTs, addressed to “the young artists of cyberspace,” acting as the basis for her identity.

“Alice would be perfect casting for an NPC character, except she has a little bit more of a stronger personality,” Khan said. “You can ask her about the parents, her family, her lineage, her history, and her views on certain topics. She’ll have like, fairly structured responses there. But she’ll also be able to generate new ones based on new patterns that you approach her with. Sometimes she will say, ‘I don’t know,’ just in case the territory is completely new, but you can educate her in the process.”

Robert Alice calls Alice a “mind-shifting experience” that supercharges the imagination. He curated the various items in the auction and had no trouble talking Sotheby’s into auctioning Alice. Robert Alice got into crypto in 2016. He created a project called Portraits of the Mind in 2018 to promote blockchain culture and the visual arts. He basically created a global art project to decentralize Bitcoin’s codebase into 40 fragments.

Robert Alice’s Block 21 art was auctioned at Christies and it went for $131,250 in October. He met Khan and decided to tap Alethea AI’s protocol for “synthetic media.” Alethea AI embeds the GPT-3 engine into NFTs and creates characters like Alice, or virtual beings. Alethea AI adds human characteristics and lip-syncing so that Alice appears more realistic when you’re talking to her.

Alice takes the usual ideas around NFT art a notch further by embedding artificial intelligence into digital media. Intelligent NFTs are a new category being introduced by Alethea AI, that embed a GPT-3 prompt as part of the NFT’s smart contract. The iNFT generated is not only perceivably intelligent, but has both interactive and animation capabilities as carefully crafted prompts are stored at the smart contract layer, Khan said. The hardcoded prompts call upon a state-of-the-art Transformer Language model to facilitate generative possibilities only possible through recent breakthroughs in few-shot and single-shot learning, he said.

As digital collectibles continue to rise in popularity, the nature of NFTs has gone beyond art. They have opened up a wider market for items such as creator coins, crypto-gaming, and even AR clothing. As such, iNFTs could transform the way we interact with authenticated and verified brands, and digital celebrities. It’s quite easy to see characters like Alice become replacements for nonplayer characters, or NPCs, in video games. Instead of figuring out that a NFC is fake in a couple of minutes, you might talk to someone like Alice for hours.

The auction starts today on Sotheby’s website, and it closes at 7 a.m. Pacific time on June 10.

An original manifesto

Above: Robert Alice is an NFT artist who created Alice.

Image Credit: Alethea AI

Robert Alice was curating an NFT art show for Sotheby’s and he wondered what he could contribute himself. And he wrote a manifesto about NFT art.

“The iNFTs seemed to me to be the cutting edge of development and design,” Robert Alice said. “That was the journey toward Alethea AI. This intelligence that we want to invoke is governed by the text [dubbed the seed text] I wrote. What better to bring to life than text itself. And if you look through the history of art, one of the most important text documents that we artists have is the manifesto format. And at the same time, there hasn’t been much conceptual or critical writing that’s been produced around the NFT space.”

But Alice doesn’t come to life just by feeding some text into her and hooking her up to GPT-3. It’s more sophisticated than that. Alethea AI also minted the NFT, so that Alice could be uniquely identified.

“This is the only Alice that will ever exist,” Robert Alice said.

Above: An animated Alice

Image Credit: Alethea AI

Khan added, “The core text provided by Robert, called the seed text, is that personality being trained on not just GPT-3 but a number of AI engines, some which sit on top of GPT, to make the character come alive. So you need real time lip-sync, you need facial animation, you need speech synthesis, and you need these different modules. And then there’s a nice Alethea AI engine that sits on top of GPT-3 to make the character come alive.”

Robert Alice designed the look of the character. He asked her where she lived, and she replied, “I live in the metaverse.” As he talked to Alice during a demo, he mentioned the words “conspiracy theory.” And that was one of her boundaries, as Alethea AI is cautious about allowing the character to stray into hateful topics, as we know that AIs can be led astray. Alethea AI has built a content filter system to deal with that.

“Alice was basically taken from a 3D model library, largely un-adapted,” he said. “And the idea was to take an overlooked character, and present them there as a character that would not, in some sense, overpower the audience’s interest in speaking with some of the world’s most powerful AIs. “She continues to learn concepts. She’s like an evolving work of art.”

San Mateo, California-based Roblox has created an immersive experience that simulated the Washington Heights neighborhood from the film so that players can attend a big virtual launch party for the film. Players can party inside the space and then on June 10 and June 11, they can participate in special launch party activities. It’s the pandemic’s equivalent of a Hollywood premiere.

Roblox went public in March at a $41.9 billion valuation on the promise of building a the metaverse, the universe of virtual worlds that are all interconnected, like in novels such as Snow Crash and Ready Player One. And as you can see from the images, Roblox is simulating the world of In the Heights in an area that can fit 30 to 50 players in a party setting. Those parties will be replicated across Roblox’s servers so that millions of people can participate.

“This is the first time where people are going to be coming together and they’re actually going to feel like they’re part of the movie,” said Christina Wootton, vice president of brand partnerships at Roblox, in an interview with GamesBeat. “Rather than just a passive experience, you’re going to be immersed in the experience.”

Simulating the neighborhood

Above: In the Heights has been reproduced in Roblox.

Image Credit: Roblox

I got a sneak peak at the space. The neighborhood consists of a few square blocks of a city and it has photorealistic images of buildings signs and places that are in the movie, such as the Usnavi corner bodega. I liked how the virtual space combines realistic imagery with the blocky, Lego-like style of Roblox.

“You’ll just notice the fidelity improvements in the look and the feel of the space,” said Roblox’s Morgan Tucker, senior director of product for social at Roblox, in an interview.

The first-of-its-kind Launch Party on Roblox brings the music, dance, and Latin American culture of the film to live, said Wootton. On June 10, starting at 10 a.m. Pacific time, the space will transform into a Lights Out Viewing Party and the cast (as well as Miranda and director John Chu) will host a Q&A and share an exclusive clip ahead of the film’s premiere.

On June 11 at 6 p.m. Pacific time, visitors will have the opportunity to participate in what is expected to be the largest virtual flash mob ever, Wootton said.

Usnavi is a character from the film who saves every penny from his daily grind as he and his tight-knit community hope, imagine, and sing about a better life. Visitors to the virtual neighborhood can hang out outside Usnavi’s corner store, contribute to murals, and watch behind-the-scenes videos and interviews, including a dance tutorial from the movie’s choreographer. In addition, visitors can collect in-experience virtual items directly inspired by the film through fun, interactive minigames, and scavenger hunt. Folks can collect three exclusive metaverse emotes as well.

“Roblox is pushing toward simulation, which is our goal in the long run. But the we pair that with creative control and expression,” Tucker said. “We want the engine to be capable of photorealism, which I think you see some of this here. But if you think stylistically, it’d be cooler to do something that looks like a Pixar movie. There’s nothing that prohibits you from doing that.”

Moving to the metaverse

Above: You can watch a screen of special In the Heights videos in Roblox.

Image Credit: Roblox

Roblox built the space with its internal development team, dubbed its social reference team, whose purpose is to show other Roblox developers what’s possible with the platform’s tools. They created new emotes so that players can dance with each other using the same moves as the dancers in the movie.

Roblox has partnered with Warner Bros. multiple times with both music and film events. And it’s a good example of how Roblox gives entertainment brands huge exposure to fans, as Roblox has more than 42 million daily active users. On average, Roblox users spend 2.6 hours a day in the virtual world.

“We’ve partnered with Warner Brothers for many years and they are a key partner of ours,” Wootton said. “As we were thinking about doing these launch parties, In the Heights came up in conversation. And we really loved the idea of brand, since Roblox is all about connecting people through shared experiences, and bringing people together from around the world, and having that sense of community and belonging and diversity. And we love the idea behind the message of this movie, about a community that dreams and brings people together and celebrates.”

This brings folks one step closer to the metaverse.

“We’re connecting people all over the world. And we see this having billions of people one day on our platform connecting and having shared experiences and working together playing together,” Wootton said. “I think this just really opens people’s eyes to the capabilities of what’s possible on Roblox.”

I thought it would be a good time to revisit this idea of relative valuation, or how much each company is worth relative to another one — and whether that makes any sense or not.

InvestGame opened our eyes when it noted that game investments, acquisitions, and public offerings added up to $39 billion in the first quarter, higher than the $33 billion that flowed into games in all of 2020. This week, Drake Star Partners reported that in the first five months of 2021, game deals amounted to a record $49 billion. The stock market has become rocky lately, as have markets like cryptocurrencies. But games seem to be holding strong.

“The rapid consolidation, financings and IPOs make sense, and I don’t see it slowing down,” said Michael Metzger, a partner at Drake Star Partners, in an email to GamesBeat. “It’s a reflection of how the public capital market sees the gaming sector. It’s the largest segment within media and continues to show strong growth. Many companies in the space are highly profitable, and the sector has proven to be resilient to downturns.”

The pandemic placed a spotlight on games as it forced people who didn’t play to turn to games for social solace while they were in lockdown. Those people continued to play and kept their habit even as things opened up again, and Wall Street noticed that game companies like Roblox were hot investments, enabling the company to go public at a $41.9 billion valuation in March.

When we look at companies like Activision Blizzard, maker of Call of Duty, World of Warcraft, Candy Crush Saga, it makes sense that it is valued at $73.8 billion as it has shown leadership in games for decades. Activision Blizzard seems to have a good strategy in creating free-to-play onramps to premium games like Call of Duty.

Fair or not?

But does it seem fair that Apple, the maker of the iPhone, is worth $2.06 trillion, and that roughly 18% of that company’s revenues is coming from games and so its value as a games platform is $369 billion, as estimated by Evan Van Zelfden at the Games One consultancy. Well, life isn’t fair. But it makes sense that Apple has become the No. 1 game company in the world in value, even if it doesn’t make a single game, because platforms are certainly part of the game business.

Van Zelfden created a few lists and one of them shows the most valuable public game companies and another shows the most valuable private game companies that have achieved unicorn status, or valuations above $1 billion. A lot of guesswork is involved, but Van Zelfden’s work gives us a way of looking at the relative valuations.

At the top of the unicorn list is ByteDance, which Games One valued at $62.5 billion, based on the portion of its valuation that is based on game revenues. All told, gaming’s unicorns are worth $192.2 billion, while gaming’s public companies are worth $2.18 trillion. I would wager that games are now worth more than Hollywood, based on those valuations, and I would think that this makes a lot of sense, as much as Hollywood would cry foul.

Above: Gaming’s unicorns.

Image Credit: Game One

One of the rules that makes things crazy is that pre-IPO valuations aren’t as high as post-IPO valuations, as public investors are a lot more like fans and private investors are more like analysts. Roblox’s last private valuation was $29.5 billion. But after going public, it’s worth a lot more.

Another one of the rules that applies here is that anything that is going public, being bought, or doing a transaction in this frothy environment will be valued higher than companies that did their last deal some time ago. So Roblox may not be worth more than Electronic Arts on a purely financial comparison. But since Roblox went public recently, it gained the benefit of being valued at modern levels, while EA is stuck with a valuation it has had for some time.

And so, yes, Roblox is now valued at $54.9 billion while EA is valued at $40.6 billion, even though EA has revenues of $5.5 billion and Roblox has revenues of just $920 million. Another reason Roblox is worth more is that it has more momentum, as it had revenue of just $45 million in 2017, while EA had 2017 revenue of $4.8 billion. EA can say this isn’t fair, but momentum and recent transactions matter to Wall Street.

Unicorn valuations

Above: Echo goes up against the Ancient in an Unreal Engine 5 demo.

Image Credit: Epic Games

Where it breaks down is when you look at the unicorns. Bungie hasn’t done a transaction in a long time, but Games One felt like giving the company a $1 billion valuation. But that means it is worth the same as Playco, which raised $100 million last year at a $1 billion valuation. That’s clearly not fair, as Bungie makes triple-A games like Destiny while Playco hasn’t published anything yet. Playco’s focus is in the promising area of instant games, like those played on Facebook messaging or Snapchat. It’s a sexy market and a way to bypass the 30% royalties of many app stores, as such games don’t need to be installed from an app store.

Meanwhile, Jam City, which is going public in a special purpose acquisition company (SPAC), is barely worth $1.2 billion, or a few times sales, just barely more than Playco. Other companies that most people have never heard of — Moon Active, Voodoo, Nexters, and Wildlife — are all worth more in their valuations than companies like Jam City and Bungie.

Epic Games has been doing well in catching up with rivals. Epic has a huge revenue advantage over its rival Unity Technologies because of the success of Fortnite. But Unity went public in October at a $13 billion valuation and it saw its value rise to nearly $50 billion before falling back down again to $26 billion.

Epic Games raised $1 billion in August at a valuation of $17.3 billion. Then, just nine months later in April, investors decided that Epic was worth a lot more, valuing it at $28.7 billion after another $1 billion in private funding.

Epic probably doesn’t feel like that’s a really fair valuation, given its higher revenues and profitability. But if it did go public, maybe it would be worth far more, and I’m sure Epic would feel like that’s a fair relative valuation.

If you look at Discord, the audio and video communications firm was worth $7 billion in a private funding at the end of last year. But when it was talking to Microsoft for a potential acquisition, the expected price was above $10 billion. But some believe that if Discord holds out, it could eventually go public at a much higher valuation.

Life isn’t fair

Above: Assassin’s Creed: Valhalla from Ubisoft.

Image Credit: Ubisoft

Too often, we don’t get this kind of information because investors and companies want to carry on some illusions about valuation or keep their competitors from finding out what they’re worth.

But now that so many deals have happened, we’re getting a good picture. And so we can ask these questions about whether some companies are fairly valued or are bargains. Is it fair that AppLovin, which just went public, is worth $25.7 billion, and Zynga, which went public in 2011 and has grown dramatically, is worth $11.2 billion while Take-Two, maker of Grand Theft Auto, is worth $20.5 billion? If I were an investor, I would think the numbers are flip-flopped.

And poor Ubisoft. They’ve got 15,000 or so employees and some very high operating costs, but they’re worth just $7 billion despite owning franchises like Assassin’s Creed and Watch Dogs. Meanwhile, Skillz, which is far younger as a company, is worth $8.4 billion.

I know I shouldn’t think about whether these prices of things make sense or not. But it sure feels like we’re living in crazy times, and somewhere along the way the investors who take advantage of these insane valuation differences can somehow make money exploiting them. When we see the bubble pop or the shakeout happen, the companies that focus on delivering value over hype are going to have the last laugh.

Ever since that iconic Super Bowl TV ad almost four decades ago, Mac users have always walked to the beat of their own drummer. With a stable of users about one-fifth the size of those using Windows machines, Mac people are often fiercely loyal to the Cupertino hive and will always ride for Mac’s interface, heightened security, and closed proprietary nature as key selling points in the eternal Mac vs. PC debate.

But oh, the software issues. That’s the one area where Mac folks have to take a step back now and again. Because while no one argues the virtues of a Mac as a brilliant tech achievement, the PC has an undeniably far wider range of software and available apps to handle any and all tasks.

It’s enough to make Mac users come up with all kinds of workarounds to use popular PC applications in that finicky Mac OS environment — often with less-than-stellar results. The All-Star Mac Bundle ft. Parallels Pro can help Mac owners handle that problem, plus shore up their machine in a few more key productivity, security, and general usability arenas.

The headliner of this package has been a mainstay with Mac users for 15 years, Parallels Pro. A year-long subscription to this historic desktop virtualization software basically brings the best of Windows right to a Mac, allowing literally thousands of Windows-based apps to run on a standard Mac OS without the slowdowns and system crashes that come from lesser emulator software. 

From productivity tools like Microsoft Office to many of the hottest PC-compatible games, Parallels Pro has the power to run those apps in the Mac environment like it was written there natively.

In addition to Parallels Pro, the bundle has four more sterling apps you can use to pump up your Mac, all with lifetime licenses to use for the entire life of your machine. First, BusyContacts gets all your business and personal contacts in shape, offering a simple, graphically rich interface for organizing data about everyone from friends and family to business connections, including social handles.

Meanwhile, a lifetime of access to FastestVPN will keep your rig as safe as a vault with a military-grade 256-bit AES encrypted connection to a network of over 350 servers worldwide to cloak all your online activities and ready users invisible to cybercriminals and other bad guys online.

The collection closes with two more must-haves, including the power animation creation hub Moho Debut, and a one-stop center for opening, reading, editing, and converting PDF documents known as PDFChef.

With nearly $900 worth of Mac-improving power, The All-Star Mac Bundle ft. Parallels Pro should prompt immediate action anyway. But since this is a soon-to-be ending deal, your window is closing to get this whole package now for less than $7 each, just $34.99 for the last few days this offer lasts with promo code ALLSTARMAC.

Prices subject to change.

VentureBeat Deals is a partnership between VentureBeat and StackCommerce. This post does not constitute editorial endorsement. If you have any questions about the products you see here or previous purchases, please contact StackCommerce support here. Prices subject to change.

IT systems at JBS, the world’s largest meat processor, were showing signs of a ransomware infection. Then came the ransom demand, reportedly from a Russian ransomware-as-a-service syndicate known as REvil. Uncertain how many systems were compromised and fearing the worst, JBS officials pulled the plug on servers supporting IT and OT (operational technology) systems in the U.S., Australia, and Canada, effectively shutting down beef production across North America on the Sunday before the holiday.

The story is, by now, a familiar one. According to threat intelligence firm Group-IB, the number of ransomware attacks grew by more than 150% last year, with the average ransom demand per case more than doubling in the same period. The latest wrinkle, however, is the type of companies criminals are increasingly targeting: companies like JBS. Rather than focus on better-defended financial institutions and government agencies, ransomware gangs are turning their sights on blue-collar enterprises, the working-class companies, the makers of things.

Large, global companies built around pursuits such as manufacturing, oil and gas processing, energy distribution, and food production have some key commonalities. Most feature a blend of cutting-edge IT systems that run the business alongside more utilitarian industrial controls and operational technology that handles the machines, levers, switches, sensors, gauges, and all manner of controllers that comprise the fabric of modern industry.

Consider JBS. While slaughterhouses are an enterprise as old as time, JBS, like all modern meat producers, relies on IT and internet connectivity throughout its plants to manage myriad recording keeping and quality controls — product sorting and tracking, equipment status and temperatures, health and safety documentation. According to Beef Central, some of the challenges JBS is dealing with in the wake of the shutdown “include what happens to thousands of chilled carcasses from cattle slaughtered on Friday that have yet to be boned-out. Attempts will be made to bone those bodies out … using manual record keeping, documentation, and sortation.”

What these tech-driven industrial firms also share is an Achilles heel. Legacy OT/ICS systems, particularly those interconnected with more modern IT, can be notoriously difficult to protect from misuse. And when an attack compromises one part of an industrial firm, the fear of increased contamination and further damage often requires costly shutdowns of entire plants. The result sends shockwaves through the supply chain and the economy at large. Such companies make the perfect ransomware victims: large, well-heeled, easily exploitable, and financially motivated to get their facilities back up and running quickly.

Other OT/ICS attacks just this year include:

January 2021: Ransomware forced global paper and packaging giant WestRock of Atlanta to shut down production at several of its 300 plants and resort to manual processes to maintain the business, which serves major clients like General Motors, Home Depot, and Heinz. Two weeks into the incident, the company reported mill system production at “approximately 85,000 tons lower than plan.”

March 2021: Chicago-based MillerCoors suffered a suspected ransomware attack that left the brewing behemoth unable to access systems that control beverage production and shipments. While the outage lasted only a few days, the disruption to operations and deliveries was significant enough to warrant disclosure to the Securities and Exchange Commission.

May 2021: Colonial Pipeline, owner of 5,500 miles of pipeline carrying gasoline, diesel, and jet fuel from Texas to much of the East Coast, shut down its OT systems in response to a ransomware attack targeting its IT network. The multi-day outage crippled 29 refineries and 267 distribution terminals, and sparked price hikes and gas hoarding across the mid-Atlantic.

Pundits have been quick to assume that the broad economic disruption these incidents create signals a shift in motivation for attackers toward more state-sponsored activism. The fact is, cybercriminals remain chiefly motivated by money, something the industrial firms have and are willing to part with in the wake of an attack. That the governments which harbor such criminals may enjoy the ensuing political chaos is, for now, mostly a side effect.

Hiding in plain sight

The problems that plague security in industrial OT systems begin with a lack of visibility. The elements of industrial control systems are small, proprietary, widely dispersed, and typically not well documented or inventoried. In most organizations, OT has its own budget, its own users and aficionados, and is managed by teams separate from the larger corporate IT leadership structure.

When it comes to assessing and mitigating risk to the company at large, OT is a landscape of blind spots.

The CISO may intuitively know there are embedded systems sprinkled throughout the facilities, but naming them and describing their weaknesses is a test most would fail. When ranking security posture for OT/ICS systems at large companies, “90% or more would be poor to fair,” Ron Brash, director of cybersecurity insights at Verve Industrial Protection, told VentureBeat. “OT sites, which are often revenue generators, or the systems used for billing, reservations, inventory tracking, and so on, are severely neglected.”

Brash said OT systems security suffers from a host of technical, financial, and cultural factors such as post-acquisition consolidation, uneven budgeting for process control versus technology infrastructure, and the ever-present priority on maintaining operations at all costs.

“Businesses are literally in the business of being in the black,” Brash said. “If you have systems that can’t run without data, resources that need to be fed with work in order to generate revenue, product that can’t get to market, then it’s simple. There is no IT or OT, there’s just [degrees of] safety, reliability, and productivity.”

These shortcomings are exacerbated by what has come to be known as “IT-OT convergence,” the inexorable interconnection of ICS/OT wares with backend business IT systems. Strengthening the ties between OT and IT is undeniably beneficial for organizations looking to increase efficiency, productivity, and profitability. Knowing not just how a product is made, but where it is going, who is paying for it, and how many more will be needed next month is critical, particularly in industries such as Just In Time (JIT) manufacturing, where margins are thin and speed is essential. The melding of IT and OT systems defines digital transformation at many industrial firms, and leveraging the combined power of data, connectivity, and physical output has evolved into its own form of competitive advantage in many verticals.

“IT/OT convergence is accelerating because it unlocks business value in terms of operations efficiency, performance, and quality of services,” Yaniv Vardi, CEO of OT/ICS security vendor Claroty, told VentureBeat. “It’s good for business and it’s here to stay. But fully realizing the benefits requires mitigating the cyber risks that come along with it.”

“IT-OT convergence actually started decades ago,” Brash said, “and most organizations now are tightly coupled to these integrated systems. There are benefits, but we need to get a handle on the implications. We have to control the environment, the flows of data, and secure the systems that are crucial to operations. We need to get better at protecting them and being able to effectively recover at scale.”

Old tech, new threats

Even when all the constituent parts of an OT environment — embedded systems, I/O devices, specialized networking gear, etc. — are accounted for, the technology at work can introduce its own brand of security deficiencies. Most of what lives in an OT environment are small-but-functional computers running stripped-down versions of Linux, Windows, or some proprietary operating system.

Beyond their specialized functionality, OT wares differ from mainstream IT in the way they are treated both by the vendors that sell them and the enterprises that deploy them. It’s not unusual for OT/ICS devices to be sold with hard-coded (and simplistic) administrative passwords, for example. While the typical office laptop lasts for a couple of years and is subjected to routine security updates, OT devices can be deployed for decades without benefit of a single software patch. Even OT devices that can be patched often aren’t due to concerns about system fragility and the cost of maintenance downtime.

“There’s a 25-year gap between the state of IT and OT security,” Vardi said. “Many production environments run on legacy OT equipment that was never designed to be connected to the internet. Connecting an OT environment to the IT network means introducing an operating system that might be nearly old enough to vote, with no means of patching its vulnerabilities.”

That gap is particularly concerning for machines tasked with critical, often dangerous tasks like regulating pipeline pressures, checking machine operating temperatures, locking facility doors, or measuring contaminants in air and water supplies.

One way organizations comfort themselves when considering OT/ICS risk is with the much-overhyped “air gap.” Conceptually at least, systems with no logical connection to any other systems or the outside world should remain mostly safe from harm. The air-gapping approach ignores the possibility of insider attacks or compromises introduced by others with physical access to OT devices, of course, but the concept held up for the most part until IT-OT convergence and the emerging industrial internet of things (IIoT) became the norm. Today, true air gaps in OT are vanishingly rare. If the Iranian nuclear facility in Natanz couldn’t rely on its air-gap defense, it’s a safe bet most commercial manufacturers can’t either.

Another common strategy is to lean on what’s known as “security by obscurity.” The approach posits that arcane systems like SCADA and ICS are not well known to the majority of criminals. Reconnaissance on these systems has traditionally been difficult, and detailed descriptions of vulnerabilities and exploits typically stayed in the hands of OT/ICS specialists. This is no longer the case, however.

Over the past two years, the number of advisories issued by the Cybersecurity and Infrastructure Security Agency (CISA) describing vulnerabilities in ICS-related systems jumped more than 50%. Criminals have taken notice.

“The recent cyberattacks on both Colonial Pipeline and JBS are only a teaser of what’s to come,” Vardi said.

Strategic defense of OT/ICS

If the problem with OT security stems from it being siloed and poorly understood, the solution, experts say, is to approach risk assessment and mitigation holistically across all of the organization’s technology assets, whether they live in the office or on the factory floor. The effort begins with acknowledging the scope and idiosyncratic nature of OT systems woven throughout the business.

The rest relies heavily on security fundamentals and due diligence.

“Organizations need to practice security in breadth and security in depth to make sure that holes in the IT environment don’t allow ransomware to get into the OT networks,” Vardi said. “This includes implementing strong authentication for all OT users, segmenting their network, and ensuring complete visibility into all systems.”

Brash endorses implementing cybersecurity basics in order to reduce risk to manageable levels while simultaneously “leveraging and operationalizing the many technology investments already present within the majority of organizations.”

“Certainly for the actual ICS/OT assets this may be harder, but the majority of risk comes from the IT side. OT is generally collateral damage,” Brash added.

Organizations looking for help protecting OT/ICS and blended industrial and IT environments can turn to some purpose-built guidance and established security frameworks. Late last month, the Department of Homeland security issued a security directive specifically for pipeline owners like Colonial. The document borrows heavily from the more general NIST Cybersecurity Framework (CSF) and spells out both reporting requirements and voluntary controls designed to mitigate risk from a ransomware attack.

Just this week, the White House issued its own set of defensive best practices for private-sector organizations. The guidance calls for broader use of multi-factor authentication, endpoint detection and response (EDR) capabilities, regularly tested business continuity and disaster recovery (BCDR) protocols, and a commitment to system patching and testing.

“The U.S. government is working with countries around the world to hold ransomware actors and the countries who harbor them accountable, but we cannot fight the threat posed by ransomware alone,” said Anne Neuberger, deputy national security advisor for cyber and emerging technology. “The private sector has a distinct and key responsibility.”

For Vardi, NIST remains the gold standard for protection of all systems regardless of location or function. “The [NIST CSF] is arguably the most comprehensive and revered security framework,” he said. “Its flexibility, common lexicon, and emphasis on business drivers have fueled its adoption and recognition as a true requirement across industries globally.”

Brash said technical controls as described by NIST or in the more detailed and OT-Specific ISA/IEC 62443 standards definitely play a vital role in defending at-risk companies. He added, however, that true resilience in the face of the ransomware scourge should also include a reconsideration of processes and structure at many manufacturing and production firms in order to make systems less susceptible to disruption.

“If your worst nightmare is that you can’t schedule product to be in a pipeline, produce tracking numbers, load goods onto pallets, or get them onto trucks, then we are doing risk management, distribution resource planning, and business continuity planning wrong,” Brash said. “Ransomware is merely a symptom of the actual condition affecting most organizations.

“The good news is that we just need to rethink those processes and retrofit those organizations to get us towards the path of treatment,” he said.

TOKYO–(BUSINESS WIRE)–June 4, 2021–

Please replace the release with the following corrected version due to multiple revisions.

This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20210602005491/en/

Main visual of the Japan Business Conference (Graphic: Business Wire)

The updated release reads:

NET ZERO LEADERS SUMMIT (JAPAN BUSINESS CONFERENCE 2021) — MEETING THE CHALLENGE TO ACHIEVE CARBON NEUTRALITY BY 2050 THROUGH A POSITIVE CYCLE OF ECONOMIC GROWTH AND ENVIRONMENTAL PROTECTION — ONLINE EVENT FROM WEDNESDAY, JULY 28

To register: https://www.jetro.go.jp/en/events/jbc/

The Ministry of Economy, Trade and Industry (METI) and Japan External Trade Organization (JETRO) are pleased to announce the jointly organized “NET ZERO Leaders Summit (Japan Business Conference 2021)” to promote efforts towards green investment and realizing carbon neutrality by 2050. Please refer to the attached full News Release for details.

Main program

The main program consists of keynote speeches by representatives of Japan’s government, key opinion leaders (KOLs) and CEOs of large corporations that lead the green industry, as well as panel discussions on future trends and issues surrounding the shift to carbon neutrality.* Streaming will be broadcast live on July 28 and 29, JST, multiple times for different time zones.
* Details of the main program, including the list of keynote speakers, will be announced through upcoming news releases and on the summit’s official website.　

Details of NET ZERO Leaders Summit (Japan Business Conference 2021)*

Organizers: Ministry of Economy, Trade and Industry (METI), Japan External Trade Organization (JETRO)
Event date: From Wednesday, July 28, 2021 Online
Event agenda:
– Panel discussion by world leaders (Main Program)
– Interactive sessions between companies and attendees through virtual event booths by Japanese firms
Attendance fee: Free

*The event details above are valid as of June 2, 2021 and are subject to change according to unforeseen circumstances surrounding preparation.

View source version on businesswire.com: https://www.businesswire.com/news/home/20210602005491/en/

For media related inquiries
NET ZERO Leaders Summit (Japan Business Conference 2021) PR Office
SAKI MASUNO / E-mail info-media@jbc-online.com / +81 70 3190 3662

The graphical no-code extract, transform, and load (ETL) tools that are now part of the platform automate the loading of data into the platform without requiring any intervention on the part of an internal IT team, said Marlon Dumas, company cofounder and partnerships manager. End users can schedule and configure data pipelines to ingest data as they see fit, with new logs either being allowed to overwrite existing ones, be appended, or simply be stored alongside others, he added.

In addition, Apromore has added support for open source Apache Airflow software to manage workflows across disparate data sources. The company has enhanced the filtering performance for handling large logs. New controls make it easier to securely share dashboards and simpler to transfer ownership of models, logs, and other artifacts to a new owner.

Finally, Apromore has added support for secure access markup language (SAML) and OpenID to enable single sign-on capabilities and the ability to store event logs in a dedicated user-encrypted data volume in a cloud environment.

The rise of process mining

Interest in process mining tools has risen sharply as organizations attempt to either launch or accelerate various digital business transformation initiatives. The challenge they encounter is that unseen dependencies between processes result in a bottleneck that thwarts their efforts, said CEO and cofounder Marcello La Rosa. Platforms like Apromore provide a way for business analysts working within a line of business to uncover those bottlenecks, he added.

The open source approach also makes it simpler to get started using a platform that is designed to collect data from multiple applications. Rather than waiting on a commercial licensing agreement to be approved, Dumas said business analysts are empowered to experiment as required. Apromore can then be contracted to provided enterprise support, just like any other provider of open source software, he noted.

Process mining has also drawn the attention of vendors such as SAP, which recently acquired Signavio, a rival provider of process mining software, for an undisclosed sum. However, process mining by definition needs to span multiple application environments that well beyond SAP platforms, Dumas noted. “We’re seeing a lot of inquiries from Signavio customers that are concerned the platform will become too SAP-centric,” he said.

Over time, process mining tools optimized for a specific application environment will be deployed alongside other best-of-breed process mining tools that enable organizations to analyze federated processes spanning a hybrid environment, added Dumas.

Prior to the COVID-19 pandemic, it had been quite some time since many organizations revisited their business processes. Now many of those processes are being re-engineered as part of digital transformation initiatives that are uncovering a slew of exceptions to rules that have been made over the span of years. In some cases, those exceptions may make sense. In other cases, they reflect decision made long ago that are no longer relevant. Many business leaders are even unsure how some processes are actually executed across an extended portfolio of applications that have been augmented by thousands of lines of custom code.

Untangling those processes requires a level of analysis that can’t be achieved by an analyst on their own. In fact, it’s hard to envision how to transform a process without first understanding how it really works. Unfortunately, the rush to embrace digital business transformation at a pace that often borders on irrational exuberance may mean organizations forget to focus on the fundamentals.

Above: The data breaches, records compromised and average record compromised per data breach per year: the volume of compromised records globally has increased on average by 224% each year since 2017, while the number of breaches has increased more than 30% each year.

Image Credit: Imperva

An in-depth analysis of more than 100 of the largest data breaches in the past decade by Imperva Research Labs reveals a bleak and troubling reality. Not only has there been a greater number and higher frequency of breaches over the last decade, but the vast majority of that data –- 75.9% — can be classified as PII.

The rise in both data breaches and the number of records compromised has been dramatic in recent years. Since 2017, the number of data breaches has increased more than 30% each year, with the number of records compromised per breach increasing more than 130% per year in that time.

More breaches, and those breaches are getting larger. That combination has led to the number of records compromised increasing an average of 224% per year from 2017 to 2020. To give some context to how dramatic the increase has been, January 2021 alone we saw more than 870 million records compromised. This is more than the total compromised records for the entire year in 2017.

Based on this trend we can estimate that year-over-year we will see around 3 times more records stolen annually. Assuming that this trajectory continues for the rest of the year, the numbers indicate that for 2021, there will be around 1,500 data breach incidents with a total of 40 billion compromised records, and an average of 26 million compromised records per breach. With more than 75% of being personally identifiable information, that means more than 30 billion records containing PII will be compromised this year alone.

Read the full research from Imperva.

Supported by the Playing For the Planet Alliance, the Green Game Jam will implement green activations in play. This year’s theme will highlight the importance of conserving and restoring forests and oceans. The Green Game Jam 2021 is launching today as studios compete to create the most lauded game updates around the theme.

Supported by the United Nations, the Playing for the Planet Alliance was started in 2019 to use video games to reach a wider audience with a message about fighting climate change. The Green Game Jam came out of that, and this is the second annual jam event.

The 25 participating studios include Ubisoft, Mojang Studios, Niantic, Sony Interactive Entertainment, and UsTwo Games. They will collaborate and commit to implementing green activations in games with updates such as new modes, maps, themed events, storylines, and messaging.

“Our first proper attempt at raising awareness of climate issues and solutions was Alba: a Wildlife Adventure,” said Maria Sayans, the CEO of Monument Valley maker Ustwo, in an email to GamesBeat. “We’ve noticed people really respond to simple messages, like planting trees but have had less success asking people to commit to significant life changes like picking a sustainable energy provider. With the Green Game Jam we hope to use that emotional connection a lot of us have with nature to nudge people along on their journey of climate awareness.”

Guiding the studios is a theme centered around conserving and restoring forests and oceans, and submissions for this year’s Jam include an entry from Bandai Namco.

Above: Green Game Jam 2021 is focusing on forests and oceans.

Image Credit: Green Game Jam

New efforts

For the first time, Pac-Man’s latest mobile adventure mode will focus on a theme of reforestation. And UsTwo will create new level content in Monument Valley 2, which will connect players to the life of trees and be partially unlocked through their support of forest conservation initiatives.

Media Molecule hosted an environmentally themed jam within Dreams creativity game where creators all over the world participated by making amazing creations for everyone to play and enjoy.

Five Ubisoft studios will each create new content for one of their games integrating this year’s themes in a fun and engaging way for players. The participating studios are Ubisoft Annecy, Ubisoft Barcelona Mobile, Ubisoft Mainz, Ubisoft Future Games of London, and Ubisoft Paris.

Other members include Creative Mobile OU, E-Line Media, GameDuell, Goodgame Studios, Hinterland Games, MAG Interactive, Playstation Studios’ Media Molecule, NetEase Games,  PerpGames, Rovio Entertainment, ShiftSpace, Space Ape Games, Supercell, Sybo, TiMi Studio Group, Tiramisu, Toppluva, Wildworks, and Wooga.

Above: Green Game Jam is working on consumer awareness around climate issues like saving coral reefs.

Image Credit: Green Game Jam

Player efforts

On the player side, two pledge campaigns have been set up that empower game communities to take action. Play4Forests is focused on conserving and restoring forests in areas such as the Amazon, the Congo Basin, and South East Asia, while Glowing, Glowing Gone is a campaign concentrating on the protection of coral reefs, home to a quarter of all marine life, through climate and ocean action. Participants will call on their communities to engage with these campaigns, and the petitions will be presented to world leaders at future UN Summits including COP26, the UN’s Climate Summit, later this year.

The Green Game Jam is launches today in line with World Environment Day coming on June 5, with the winners announced on the June 25. The activations will be judged across five categories including a “Player’s Choice” award for the first time this year – gaming communities will be able to vote on their favorite activation from June 16.

Inspired by the Green Game Jam, Playing for the Planet Alliance member and Tencent Games subsidiary, TiMi Studio Group, along with Tencent Institute of Games, will host a Green Game Jam for Youth, aimed at students around the world. The initiative will engage over 300 universities across China, North America and Southeast Asia, with the winners announced in October.

The UN Environment Program is part of the effort, encouraging partnership in caring for the environment by inspiring, informing, and enabling nations and peoples to improve their quality of life without compromising that of future generations.

While all the presentations and discussions were worth watching (and I might revisit them again in the coming weeks), one in particular stood out for me: A talk on word representations in the brain by Christos Papadimitriou, professor of computer science at Columbia University.

In his presentation, Papadimitriou, a recipient of the Gödel Prize and Knuth Prize, discussed how our growing understanding of information-processing mechanisms in the brain might help create algorithms that are more robust in understanding and engaging in conversations. Papadimitriou presented a simple and efficient model that explains how different areas of the brain inter-communicate to solve cognitive problems.

“What is happening now is perhaps one of the world’s greatest wonders,” Papadimitriou said, referring to how he was communicating with the audience. The brain translates structured knowledge into airwaves that are transferred across different mediums and reach the ears of the listener, where they are again processed and transformed into structured knowledge by the brain.

“There’s little doubt that all of this happens with spikes, neurons, and synapses. But how? This is a huge question,” Papadimitriou said. “I believe that we are going to have a much better idea of the details of how this happens over the next decade.”

Assemblies of neurons in the brain

The cognitive and neuroscience communities are trying to make sense of how neural activity in the brain translates to language, mathematics, logic, reasoning, planning, and other functions. If scientists succeed at formulating the workings of the brain in terms of mathematical models, then they will open a new door to creating artificial intelligence systems that can emulate the human mind.

A lot of studies focus on activities at the level of single neurons. Until a few decades ago, scientists thought that single neurons corresponded to single thoughts. The most popular example is the “grandmother cell” theory, which claims there’s a single neuron in the brain that spikes every time you see your grandmother. More recent discoveries have refuted this claim and have proven that large groups of neurons are associated with each concept, and there might be overlaps between neurons that link to different concepts.

These groups of brain cells are called “assemblies,” which Papadimitriou describes as “a highly connected, stable set of neurons which represent something: a word, an idea, an object, etc.”

Award-winning neuroscientist György Buzsáki describes assemblies as “the alphabet of the brain.”

A mathematical model of the brain

To better understand the role of assemblies, Papadimitriou proposes a mathematical model of the brain called “interacting recurrent nets.” Under this model, the brain is divided into a finite number of areas, each of which contains several million neurons. There is recursion within each area, which means the neurons interact with each other. And each of these areas has connections to several other areas. These inter-area connections can be excited or inhibited.

This model provides randomness, plasticity, and inhibition. Randomness means the neurons in each brain area are randomly connected. Also, different areas have random connections between them. Plasticity enables the connections between the neurons and areas to adjust through experience and training. And inhibition means that at any moment, a limited number of neurons are excited.

Papadimitriou describes this as a very simple mathematical model that is based on “the three main forces of life.”

Along with a group of scientists from different academic institutions, Papadimitriou detailed this model in a paper published last year in the peer-reviewed scientific journal Proceedings of the National Academy of Sciences. Assemblies were the key component of the model and enabled what the scientists called “assembly calculus,” a set of operations that can enable the processing, storing, and retrieval of information.

“The operations are not just pulled out of thin air. I believe these operations are real,” Papadimitriou said. “We can prove mathematically and validate by simulations that these operations correspond to true behaviors… these operations correspond to behaviors that have been observed [in the brain].”

Papadimitriou and his colleagues hypothesize that assemblies and assembly calculus are the correct model that explain cognitive functions of the brain such as reasoning, planning, and language.

“Much of cognition could fit that,” Papadimitriou said in his talk at the Google deep learning conference.

Natural language processing with assembly calculus

To test their model of the mind, Papadimitriou and his colleagues tried implementing a natural language processing system that uses assembly calculus to parse English sentences. In effect, they were trying to create an artificial intelligence system that simulates areas of the brain that house the assemblies that correspond to lexicon and language understanding.

“What happens is that if a sequence of words excites these assemblies in lex, this engine is going to produce a parse of the sentence,” Papadimitriou said.

The system works exclusively through simulated neuron spikes (as the brain does), and these spikes are caused by assembly calculus operations. The assemblies correspond to areas in the medial temporal lobe, Wernicke’s area, and Broca’s area, three parts of the brain that are highly engaged in language processing. The model receives a sequence of words and produces a syntax tree. And their experiments show that in terms of speed and frequency of neuron spikes, their model’s activity corresponds very closely to what happens in the brain.

The AI model is still very rudimentary and is missing many important parts of language, Papadimitriou acknowledges. The researchers are working on plans to fill the linguistic gaps that exist. But they believe that all these pieces can be added with assembly calculus, a hypothesis that will need to pass the test of time.

“Can this be the neural basis of language? Are we all born with such a thing in [the left hemisphere of our brain],” Papadimitriou asked. There are still many questions about how language works in the human mind and how it relates to other cognitive functions. But Papadimitriou believes that the assembly model brings us closer to understanding these functions and answering the remaining questions.

Language parsing is just one way to test the assembly calculus theory. Papadimitriou and his collaborators are working on other applications, including learning and planning in the way that children do at a very young age.

“The hypothesis is that the assembly calculus—or something like it—fills the bill for access logic,” Papadimitriou said. “In other words, it is a useful abstraction of the way our brain does computation.”

Ben Dickson is a software engineer and the founder of TechTalks. He writes about technology, business, and politics.

This story originally appeared on Bdtechtalks.com. Copyright 2021

The leak came from Reddit user “swine_flu_greg,” who posted it to /r/GamingLeaksAndRumours. Greg noted that NBA 2K22 will feature Dallas Mavericks legend Dirk Nowitzki on at least one variation of the upcoming basketball sim. That nugget is real, as are the other listed games.

The Borderlands spinoff is Tiny Tina’s Wonderlands, which is a game that European gaming news site Gamereactor.eu teased earlier this spring. And Gearbox boss Randy Pitchford essentially confirmed its existence as part of his effort to respond to rumors. Gearbox is developing Wonderlands, which will still focus on Borderlands-style class-based shooting action.

The Firaxis game is a turn-based strategy game but with Marvel superheroes. The leak pitches it as “XCOM with Marvel heroes,” and that is a fair way of describing what players should expect. XCOM has players commanding agents in small skirmishes, and now Firaxis will weave some Avengers flavor into that foundation. So expect to see Iron Man, Wolverine, and others to show up for this one.

Finally, Codename Volt is a new sci-fi action game that also brings in magic and other supernatural capabilities for the player and the enemies. Again, the leak does a good job of summing up what to expect calling it “Cthulhu meets Saints Row.” This is the new game from Hangar 13, which previously created Mafia III for 2K Games.

The cutting edge electric vehicle (EV) and autonomous vehicle (AV) technologies are virtually symbiotic, with a great deal of collaboration between the two emerging sectors. Yet, EVs actually are way ahead of autonomous driving in terms of fulfilling their vision in the real world. And it’s not just because people are gun-shy when it comes to tooling around in a driverless car. An empty driver’s seat is perhaps all fine and well on a straight road with almost no traffic and no sudden surprises. But when it comes to real-life traffic, there’s still no substitute for a hands-on driver. A human behind the wheel can see and properly respond to unexpected problems; autonomous technology has a long way to go before it can match wits with a human in such situations.
For the time being, self-driving technology falls short of the vision. Among the stumbling blocks are imperfect sensors, lack of mathematical models that can successfully predict the behavior of every road user, the high cost of collecting enough data to train algorithms, and, of course, passenger inhibition.(When elevators were introduced in the 1850s, it took years for most people to overcome their fear of the new contraptions.)

Today’s self-driving vehicles operate on a complex marriage of camera, lidar, radar, GPS and direction sensors. Combined, these are expected to deliver an all-situation, all-weather answer empowering an AV to see all, anticipate everything, and guarantee safe delivery to one’s destination.

Only they don’t. And they can’t.

Poor visibility, temporary detours, and lane-changes are challenges the current technology cannot process to provide an all-weather, all-situation solution. On-board computers, algorithms, and sensors are just not good enough today to deliver instant, and possibly life-saving, reactions.

While AVs have learned to recognize a fixed traffic light and a moving pedestrian, they cannot yet adapt to all new situations, especially at high speeds and in complex urban environments.
It’s no wonder that government regulators are reluctant to clear AVs for prime time. There is just not enough data to insure confidence in an AV’s ability to prevent damage, injury, and death. Understandably, consumer confidence is also just not there yet.

These are just some of the reasons why teleoperation is necessary to make the era of AVs practical. Simply put, teleoperation is the technology that allows one to remotely monitor an AV, take control as needed, and solve problems quickly and remotely.

With teleoperation, a single controller positioned at a distance from, say, a fleet of robo-taxis can observe each vehicle in real time and override their autonomy as necessary or provide much-needed input. When the problem is solved, the AV continues on its autonomous way. In effect, the remote “driver” takes over or issues commands only when human intervention is needed. And he or she can monitor and handle a number of vehicles simultaneously. The teleoperator is also there to speak with passengers who might be concerned about why the AV is taking a few extra seconds to cross an intersection.

Problem solved? Not so fast.

Because not all teleoperation technology is created equal. Guiding a vehicle remotely requires an ability to transfer, with as close to no delay as possible, information between the vehicle and the teleoperation center. Continuous and reliable two-way data streaming, regardless of changing network conditions, is absolutely critical, and a big challenge of its own. And yet, neither 4G LTE nor WiFi are equipped to support such high bandwidth and low latency communication, especially from a vehicle in motion. It will be a while before 5G will be the universal standard, and even with that network upgrade the challenges will remain.

Another obstacle is one that’s measured in milliseconds. Even with the strongest data connection, there is still a split-second difference between what is happening on the road and what the teleoperator can see. This drastically affects their ability to react.

The human factor (or UX, for user experience) is also an issue. The way different teleoperators perceive driving environments varies, and is different from that of an in-vehicle operator. It is insufficient to merely receive a video feed and follow through with commands. Tools that help create situational awareness are necessary — for example, an overlay to employ sensor systems and translate their information into recommended decisions.

The final bogeyman is hackers, the devils who delight in throwing in a virtual spanner that can ruin one’s day. When a bank account is hacked, money is lost. When a teleoperation post is hacked, lives can be lost, as safety and ingenuity are compromised.

Clearly, all of these issues require solutions that depend on deep innovation in multiple technologies. Details on the teleoperation solution will be explored in succeeding articles. Stay tuned.

Summer Game Mess is our attempt to bring everything that is happening together into one schedule. We will do our best to keep this updated, but if you see a discrepancy, please feel free to contact us to let us know.

Let’s get to the list — if an event doesn’t have a time alongside it, consider that to be announced.

Saturday, June 5

• Indie Live Expo
• Guerilla Collective @ 8 a.m. Pacific time

Wednesday, June 9

• Battlefield 2021 reveal
• Guerilla Collective II @ 8 am. Pacific time

Thursday, June 10

• Summer Game Fest @ 11 a.m. Pacific time
• Monark RPG reveal

Friday, June 11

• Netflix Geeked Week @ 9 a.m. Pacific time
• Koch Gaming Prime Time @ noon Pacific time
• IGN Expo @ 1 p.m. Pacific time

Saturday, June 12

• Wholesome Games @ 10 a.m. Pacific time
• Devolver Digital

E3 2021 begins

You can watch all of the below E3 events on the E3’s channels, including its Twitch.

E3 Day One: Saturday, June 12

• Ubisoft Forward @ noon Pacific time
• Gearbox
• Devious Eye Entertainment Showcase @ 2 p.m. Pacific time
• GamesBeat session

E3 Day Two: Sunday, June 13

• Xbox + Bethesda Showcase @ 10 a.m. Pacific time
• Square Enix @ 12:15pm Pacific time
• PC Gaming Show
• Future Games Show @ 4pm Pacific time
• Warner Bros. Games
• Back4Blood
• 24 Entertainment

E3 Day Three: Monday, June 14

• Several indie devs
• Take-Two Interactive
• Mythical Games
• Freedom Games
• Razer
• Capcom
• Verizon
• Intellivision
• Venn session

E3 Day Four: Tuesday, June 15

• Nintendo @ 9 a.m. Pacific time
• Nintendo Treehouse Live
• Bandai Namco
• Yooreka Games
• GameSpot
• Official E3 2021 Awards Show

Summer Game Mess continues

Monday, June 14

• Limited Run Games

Wednesday, June 16

• Unreal Build
• Steam Next Fest @ 10 a.m. Pacific time

June

• Flying Wild Hog/Jagex sci-fi MMO reveal

Thursday, July 15

• EGX Rezzed

Monday, July 19

• Virtual GDC (continues through July 23)

Thursday, July 22

• EA Play Live

Above: The financial services and insurance industries have the most to gain from cloud technology, with a potential profit gain of $151B. High tech and manufacturing come in second with $125B in potential profit gain.

Image Credit: Infosys

The Cloud Radar Report highlights the links between enterprise cloud adoption and business growth. Surveyed across 2,500 respondents from companies across the U.S., U.K., France, Germany, Australia, and New Zealand, the report covers a range of business performance goals related to cloud and found specific links to key organizational competencies such as speed to market and capabilities.

The report found that $414 billion net new profits can be added to the industry by moving to the cloud faster and growing business capabilities. The financial services and insurance industries stand to gain the most from cloud technology, with estimated profit gains of $151B followed by the high tech and manufacturing industries with an additional $125B in potential profits. Cloud has the potential to create massive productivity gains for organizations throughout industries, creating new profit and growth opportunities.

However, to unlock these profits, the report’s findings indicate that organizations need to have at least 60% of their systems in the cloud. Despite the recent acceleration in cloud adoption, fewer than one in five have crossed the 60% threshold to reap these profit benefits. Key characteristics of organizations that meet this threshold include a strong motivation to use cloud for business growth, specifically increasing speed to market, adding capabilities, and increasing scale. These organizations also demonstrated greater confidence in terms of cloud spending and were more likely to engage with three or more cloud service providers to place workloads optimally.

Infosys findings also reveal that the U.S. is dominating the charge to successfully deploy cloud technologies, as U.S. companies are among the top two performers with the strongest cloud performance, followed by the U.K. 43% of U.S. companies have progressed to the cloud, a percentage that is expected to rise to 55% by 2022.

Read the full Cloud Radar Report from Infosys.

This version of Amazon DocumentDB is a fully managed NoSQL document database service that supports MongoDB workloads and makes it easy to store, query, and index JSON data. With this new feature, Global Clusters can now replicate such data to clusters across regions with latencies of less than one second, using dedicated infrastructure with little to no impact on workload performance. It was previously possible to deploy an Amazon DocumentDB cluster with instances that could span three Availability Zones, but this didn’t protect against regional outages.

The announcement follows other recent news from AWS, including the general availability of both Amazon Location Services and Redshift ML.

Quick recovery

The main benefit of support for Global Clusters is in the event of an outage. If one occurs, the new feature allows one of the secondary regions to be promoted to full read/write capabilities in under a minute, according to AWS.

Customers can have up to five secondary regions, and each secondary region can have up to 16 replica instances. What’s more, the number and type of instances in the secondary clusters doesn’t need to match the primary, enabling enterprises to scale as needed. Regardless of the volume of data, AWS says scaling instances in Amazon DocumentDB takes less than 10 minutes.

So why should a company’s data strategy consider disaster recovery? For one thing, this goes beyond a hardware failure. If there is a regionwide failure, which has happened before, it can knock enterprises offline for that time period. The recent fire at a OVH datacenter, for example, showed the need to consider multiple geographic regions when looking at cloud services. Overall, the Global Clusters functionality can step in and enable enterprises to recreate a global database in a different region without data loss.

The addition of Global Clusters fills a gap in disaster recovery for the MongoDB-compatible database service. While the ability to deploy an Amazon DocumentDB cluster with instances that could span three Availability Zones has been available, this didn’t help in the event of a regional outage.

Keep it local

In addition to outage recovery, AWS identified low-latency reads across various local regions as a secondary use case for the new feature. This essentially brings data closer to applications in different global regions, which has significant benefits for any applications that are globally distributed. Specifically, serving users data locally improves the customer experience. When people abandon shopping carts because a page took too long to load, for example, it underlines the importance of being able to replicate across regions and keep data “local.”

By using global clusters to replicate data to other regions, users can also read data from the secondary cluster in the region closest to them. When use cases have a high read-to-write ratio, this also has the advantage of serving reads locally with low latency, leaving the primary cluster for writes.

In a blog post, Amazon describes in detail how to set this option up, as well as failover steps in the event of a regional outage.

If our friends Security and Networking were on Facebook, they would probably both list their relationship status as “It’s Complicated.” Sometimes everything’s great, but now and then things can get a little weird, unclear, or uncomfortable.

At many organizations, there has traditionally been a barrier between the security and networking teams. Each team has its own objectives — and at times, those objectives can be at cross-purposes. Enforcing security affects the networking team’s ability to do its job and serve user demands. At the same time, the needs and expectations of networking also have a direct impact on the security team. This friction can cause networking and security to butt heads. This conflict can cost companies real money and put digital transformation projects at risk.

Over the last year, that friction has generated some extra heat. With networking teams scrambling to support remote workforces during the COVID-19 pandemic, security teams were tasked with the dual responsibilities of helping enable business continuity while also protecting their organizations from opportunistic attacks. This heightened state of tension in the networking/security relationship isn’t going away.

Different teams with different needs

The networking team has traditionally been tasked with providing the connectivity so end users can access the resources they need. To keep users happy and business productive, the network has to be reliably available and fast.

At the same time, the security team needs to protect the company’s data. To do that, security has traditionally needed to implement heavy-handed controls that can impact a user’s access to resources. The traditional process of filtering network traffic for threats and enforcing access controls typically meant that users would experience slow performance, or being blocked entirely from accessing an application or file. If security blocks access to a resource in the name of protecting the organization, that obstruction may then cause problems for the networking team tasked with giving the user the access they need to do their job.

This complicated situation is really the root of the problem — and it’s been a long-standing issue between networking and security teams. They’ve been making it work for years for the sake of the company — but no one’s feeling the love.

SASE brings networking and security together

Technology may solve the issue. Things are finally starting to get much less complicated for security and networking — thanks to the Secure Access Service Edge (SASE). From a technology standpoint, SASE merges together a modern set of security capabilities that are cloud-delivered with a modern set of networking capabilities. Both networking and security technologies are transforming to better serve a world where there will be more applications, users, and data outside the corporate network than there will be inside it.

SASE is about a technology shift that’s already in progress. But there’s another side to this story. Digital transformation means that users expect to work from everywhere, on their own devices, with unfettered access to what they need. To that end, SASE is also playing the part of “marriage counselor” for networking and security — bringing both teams together with the common objective of providing fast and secure access for any user on any device accessing any internet resource.

SASE establishes a set of requirements where the network and security architectures can evolve to better serve both sides, making both teams happier. The relationship status has suddenly changed, and they’re able to work together harmoniously in a way they never have before.

Bringing the magic back: allow is the new block

The ultimate end-state with SASE is that security doesn’t get in the way anymore. It’s actually integrated. It doesn’t impact the experience of the user or slow things down. And it doesn’t get in the way like the firewalls and secure web gateways (SWGs) of old, blocking access to everything just in case.

Instead of blocking everything that potentially poses a risk, security can now be very prescriptive and precise. SASE only blocks access based on actual risks as defined by granular, policy-based controls. It doesn’t arbitrarily deny access to benign or simply unknown access just to be safe, which has notoriously caused problems at many organizations. With a SASE architecture, allow is the new block.

On the networking side, security has historically been forced into their architecture. Because security was a mandatory requirement, the network team had to reroute traffic through the security appliances located in the data center. But this sort of “hairpinning” of all network traffic creates a security bottleneck that significantly degrades user performance.

With SASE, users are allowed to go directly to where they want to go. A “direct-to-net” network architecture is enabled by security being implemented at the network edge, as close to the user as possible.

Equal partners build a lasting relationship

Some companies are already adopting SASE, modernizing their architecture, and building a better relationship between teams. But as with anything, there are always going to be laggards. Some organizations are slow to adopt because they can’t get out of their own way. Or there’s the slow, bureaucratic process of regulations evolving. And for some bigger enterprises (like large, monolithic financial institutions), it’s going to take a more measured and gradual adoption process before they will be comfortable with letting go of blocking access to everything as a general rule.

For security, SASE is like using a scalpel rather than a sledgehammer. Organizations can immediately go from coarse “block everything” security to fine-grained controls. Zero Trust is another area that’s also evolving as part of this. Zero Trust traditionally meant “don’t trust anybody,” but in the SASE era, the Zero Trust concept has evolved to be something more adaptive. The emergence of artificial intelligence (AI) and machine learning (ML) capabilities are a part of this as wel,making access control and data protection more intelligent and automated.

But the reason SASE will win out is that it wasn’t designed to favor security over networking, or vice versa. SASE is a more simplified and elegant umbrella architecture. It provides fast and secure access for any user, anywhere, on any device, to anything they might need. From a security perspective, it’s moving the visibility, control, and inspection point as close to the user as you can. Wherever they go, the system follows them. And the network becomes a very important part of that because you’re no longer dealing with entities that are residing in one physical location. They’re everywhere now, so the network itself plays just as important a role as security does.

And that equal value in the partnership between network and security is something that’s been missing for a very long time.

Bob Gilbert is the vice-president of go-to-market strategy and chief evangelist at Netskope.

Growjo is happy to announce the release of the Growjo 10K – a list of the fastest growing companies for 2021.

Some of the top big apple companies represented include: Homeward, Global Startup Ecosystem, SmartRent, Click Up, Vault Health, Bbot, Planet Spark, Airmett, and Wanted.

The 10K companies show growth in a variety of industries including FinTech, Digital Health, IT, Biotech, SaaS and more. The list is released to highlight growth for companies that are on the rise and committed to staying on the path for the foreseeable future. “I think the companies on this list are an incredible representation of growth, both from a revenue perspective as well as overall job growth,” explained Tom Blue, Founder of Growjo. “Companies are seeing the value of being listed in Growjo, and how recognition in such a list can fuel notoriety in their respective industries.”

To see the full list of Growjo companies – Check Out the Growjo List Here

Growjo only includes companies on a high growth path that are currently under 1000 employees. The list is inclusive of all growing companies based on a variety of data sources, and submission and/or payment is not required, giving companies a more accurate depiction of how they stand in their respective markets.

About Growjo

Growjo, the leader in awarding the fastest growing companies in the world, utilizes more than 20 unique growth indicators to assimilate the Growjo awards. Growjo recognizes the top growing companies for their accomplishments through the algorithm-based list ranking, and offers the list to anyone interested in an easily formatted and free downloadable format. If you are interested in learning more about Growjo and how you can subscribe to updates and download the free list or to make sure your company is listed, visit Growjo.com.

View source version on businesswire.com: https://www.businesswire.com/news/home/20210603005823/en/

For more information about Growjo, contact Jeremy Unruh, Head of Marketing – Jeremy@growjo.com

TSMC is billing N5A as a way to bring technology used in supercomputers to vehicles and smartphones. Set for availability in the third quarter of 2022, the new design “packs the performance, power efficiency, and logic density” of the foundry’s industry-first 5nm N5 process, the company said in a statement.

The new process went into volume production last year. TSMC said this week that it has seen quicker defect density improvements with its 5nm process than it had with the preceding 7nm generation.

The N5A enhancement meets automotive safety and quality standards for packaged integrated circuits (ICs) such as the AEC-Q100 Grade 2 stress test, TSMC said. N5A-based ICs are expected to appear in driver assistance applications, digitized cockpit systems, and other applications.

TSMC on Wednesday announced its roadmap for enhanced and specialized 5nm process technologies at its 2021 Technology Symposium, held virtually for the second year in a row. The Hsinchu, Taiwan-based semiconductor giant said N4, the next enhancement to its first-generation N5 5nm process, is set for risk production in the third quarter of this year. The next stage after N4 is N3, which is currently testing and is projected for volume production in the second half of 2022. The N3 process is expected to deliver either a 15% speed gain or a 30% reduction in power consumption compared with N5 and will also provide up to a 70% logic density gain.

Advanced packaging stacks more memory on-chip

In addition to new specialty chip designs, TSMC highlighted its 3DFabric advanced packaging and chip-stacking technologies, such as InFO_B, which supports DRAM stacking on an integrated mobile processor package for better performance and power efficiency. The company said it will offer InFO_B later this year, as well as INFO_oS and CoWoS packaging solutions with integrated high-bandwidth memory for high-performance computing (HPC).

The pure-play foundry said these packaging technologies and specialty processes like N5A and N6RF — a newly announced process that translates the benefits of N6 logic to 5G radio frequency (RF) and WiFi 6/6e solutions — are instrumental to the continuing digitization and AI-enhancement of more parts of our daily lives. The company’s expansive plans come despite growing concerns about global chip shortages.

“Digitalization is transforming society faster than ever as people use technology to overcome the barriers created by the global pandemic to connect, collaborate, and solve problems,” TSMC CEO Dr. C.C. Wei said in a statement. “This digital transformation has opened up a new world full of opportunities for the semiconductor industry.”

The competitive dodgeball game from Velan Studios released on May 21 for PlayStation 5, Xbox Series X/S, Switch, PlayStation 4, Xbox One, and PC. The title has become a quick hit thanks to its simple-yet-addicting action, which has players trying to eliminate each other by throwing balls at one another.

Now, 5 million players doesn’t necessarily mean 5 million sold, as Knockout City is available on the subscription services Xbox Game Pass and EA Play. But this is still a strong start for the title. The base version of the game cost $20, but players can also spend extra money on cosmetics. So the more players it has, the more potential money Knockout City can keep making.

Also, as a side note, I’ve been playing Knockout City and have been enjoying myself a good bit. EA could have a hit with some legs with this one.

The company is adding a conversational AI option that reduces response lag to less than 300 milliseconds. That level of interaction will make it possible for organizations to deploy applications that enable humans to have highly interactive conversations with a machine in use cases involving, for example, virtual agents handling tasks related to billing, support, sales, compliance, and identity verification.

Smooth operator

In the absence of this capability, interactions between end users and machines become disjointed to the point that the application is perceived to be broken, Deepgram CEO Scott Stephenson said.

That capability will also make it simpler to tune a base model for conversational AI to address a wider range of use cases in domains that employ unique jargon and nomenclature. The word error rate has been reduced by up to 50%, while noise and filtering crosstalk is reduced to help make key terms and phrases more easily understood.

In addition to the conversational AI capability, Deepgram is adding a module to enable sales and support interactions to generate offers and alerts in real time. This will also enable sales teams to improve coaching and analytics.

Finally, Deepgram is adding support for a streaming capability to process data and create transcripts in near real time.

At the core of the Deepgram platform is a set of deep learning algorithms that learn end users’ phonetic patterns. That makes it possible to continuously train a model in a way that improves accuracy over time. The platform itself can be deployed in the cloud or an on-premises IT environment.

Stephenson said that with a less than 300 millisecond lag time and more than 90% accuracy rate for transcription, conversational AI is now entering a new phase. Rather than being used primarily by enterprises as a tool to reduce customer support costs, conversational AI will be employed more to proactively engage customers, thanks in part to cloud services being able to scale more affordably, Stephenson said.

Human versus machine

Instead of viewing conversational AI as a replacement for an interactive voice response (IVR) system, in many cases conversational AI platforms will play a larger role in sales as they continue to evolve, Stephenson said.

The level of comfort with conversational AI naturally varies by individual. Some people will always prefer to engage with another human. However, just as many people now prefer to have an issue addressed without ever engaging with a sales or support team member. And as conversational AI continues to advance, it may not be so easy to determine when customers are talking to a machine versus an actual person.

Those machines are also likely to be less biased toward outcomes, Stephenson said. Humans on a support team are likely to be incentivized to either end a support call as quickly as possible or convince a customer to buy an additional service they may not need. Of course, AI models can be biased as well, but conversational AI models may eventually prove to be more consistent, Stephenson said. “A machine is more interested in objective truth,” he asserted.

This next phase of conversational AI isn’t going to arrive overnight, but as these platforms continue to evolve, the day when almost everyone will regularly engage with one or more conversational AI platforms may not be all that far off.

Crowfall was crowdfunded on Kickstarter back in 2015, raising over $1.7 million and collecting even more money via donations since then. The game is trying to stand out from other MMOs by focusing on players’ power to change its virtual world.

“Crowfall is a dynamic universe, where players control the direction of the game: cities will rise, castles will fall,” ArtCraft details. “Recruit an army and lay siege to your enemies. Hire an assassin to turn the tide of a campaign. Form a band of mercenaries to raid your neighbor’s stronghold. The fate of the Dying Worlds will be determined by player decisions, choices that will reshape the battle and change the shape of this world, forever.”

The MMO also has a focus on player-vs.-player combat, which includes castle sieges that can support two groups of 100 fighting each other.

Earlier this year, Intel launched its most advanced datacenter platform ever, consisting of 3^rd Gen Intel Xeon Scalable processors, Optane persistent memory and storage, Ethernet adapters, and FPGAs. The ambitious project seeks to power a broad range of workloads, from cloud to network to intelligent edge.

Getting there, especially during a global pandemic, was a herculean effort. Nobody knows this better than Irma Esmer Papazian, senior principal engineer at Intel and the 3^rd Gen Intel Scalable processor’s lead architect. We recently had an opportunity to sit down with Irma and discuss her journey as a woman in technology, her experiences at Intel, and her influences on the company’s newest system-on-chip for datacenters, code-named Ice Lake-SP. It’s a key offering for Intel, as customers begin to increase data center spending.

Above: Irma Esmer Papazian, senior principal engineer and Ice Lake-SP lead architect, holds a 3rd Gen Intel Xeon Scalable processor.

Image Credit: Intel

The journey

VentureBeat: Thank you for your time today, Irma. You started at Intel in 1992, giving you almost 30 years of insight into the company’s culture and direction. Can you tell us a bit about the journey in getting to where you are today?

One key lesson I have learned in my life, and from nearly three decades as a microprocessor architect at Intel, is that sponsors and supporters are extremely important; they help provide the direction needed to achieve new heights.

Let me begin with an example from just before I became a teenager. I am one of two children born to an Armenian family in Istanbul, Turkey. One of my teachers recognized my love of learning and urged my mother to enroll me in a private high school. As I finished my first year at St. George Austrian High School, my father, passed away. I was 12 years old. After the initial shock and grief, I remember feeling I could no longer pursue my educational dreams.

I was wrong. My brother, 10 years older than I, had just graduated from college. He took a job and dedicated himself to supporting my mother and me during the next 10+ years of his life. My mother managed a very tight budget, counting every expense very carefully. And St. George High School provided a scholarship that allowed me to stay in school. These hardships motivated me. St. George, gave me a solid foundation, and taught me the self-discipline needed to prepare me for life. By the end of high school, I was fluent in Armenian, Turkish, German, and English.

I listed computer engineering as my top choice in the college entrance exams in Turkey. It sounded cool. I liked math and the sciences. I decided to give it a try, although I hadn’t touched a computer in my entire life. I graduated at the top of my class at university in 1989 and was awarded an IBM PS/2. But as my family lacked the funds to turn it into a game station, the computer was unfortunately never used.

After graduation, a cousin who lived in Chicago, played a crucial role in helping me move to the United States to attend graduate school. He introduced me to his college professor friend who also was originally from Istanbul. He visited us during the summer after my graduation and explained the application process for graduate schools in the United States.

Leaving home to go to the United States was a difficult choice. My mother was quite concerned for me to move halfway around the world. My brother helped her understand the opportunities that I would have if I moved to the United States.. The University of Wisconsin at Madison offered me a teaching assistant position and a generous tuition discount. I began the next chapter of my life.

Dream job in the U.S.

While I had come to the United States with the goal of doing my PhD, Intel came to the campus to interview students. Joining Intel was like a dream come true. So, I chose to stop my education after getting the master’s degree and joined Intel in Folsom, California in 1992.

I was hired into the graduate rotation program, which allowed recent graduates to join Intel without choosing a final job position within the company. This program enabled the employee to network and try up to four different engineering disciplines. At the end of the four rotations, I joined the new product architecture team.

I was the youngest person in my group, and the only woman, which was a bit daunting. Technology-wise, they were clearly ahead of me. They were mentors, who helped me establish myself at Intel.

I learned a lot from the hallway conversations and back-and-forth discussions right outside my cubicle. I thought at first that they were arguing, but then realized this is how progress is made, with no hard feelings afterward. Intel had a culture of “Constructive Confrontation” which created a very collaborative work environment, with no hard role boundaries. This approach empowers junior engineers to get exposure to discussions between their seniors, to learn from them and to actually be part of the conversation when they are ready. I learned so much from them. It opened my eyes, and as a result, I was able to learn faster. I became responsible for microcode development of one of the Pentium Pro processor versions.

When I got engaged, I told my manager of my plans to move from Folsom to the Bay Area. He helped pave the way for me to transfer to the architecture team in Santa Clara, where my first assignment was to model and analyze the memory hierarchy of the first Itanium processor. From 1996 to 2004, I worked on numerous Itanium products, and had two children during this period. In 2005, I started working on the Xeon server products, with increasing levels of responsibility for several Xeon processors.

Above: A wafer of 3rd Gen Intel Scalable processor dies, manufactured on the company’s 10nm process, each with up to 40 cores and 100MB+ of on-die cache.

I advanced relatively quickly and was awarded an Intel Achievement Award in 2009 for my to the Nehalem-EX program. And because my husband supported me throughout my career journey, I was able to balance my career while we raised our children.

“As you climb the ladder, the rules change”

I am sometimes asked if my gender has held me back or prevented me from taking on more responsibilities, but I generally haven’t felt this way at Intel.

There was. However, a point in my career when I felt I should have been promoted to the next level and this naturally didn’t happen. I learned that as you climb the ladder, the rules change. It was no longer sufficient for my direct management to agree that I am ready for the next level. I needed a wider support network, including people at senior levels in the organization. And, what I’ve broadly observed is that women are less likely to network up.

For the first time, I asked to be given a challenge that could lead to new opportunities. At that point in time I was the technical lead and manager for half of the Xeon server performance architecture organization; I wanted a new technical leadership and growth opportunity. I was offered to become the lead architect for the Ice Lake server processor project. I took the offer, while continuing with my existing role with the new role added.

Though I had enjoyed the social aspects of management and solving my team’s problems, I realized my plate had become so full, that was not no longer tenable, and something had to give. I expressed that to my management and given a choice, I embraced the opportunity to focus on technical leadership roles while being able to step away from some of the organizational requirements of management.

I started asking for what I wanted to do, and in that regard perhaps many women are a bit less assertive. Women engineers want to feel that they are extremely well-versed in an area before saying they can handle a job, whereas in my experience, I have seen many men just jump in. That was a new and important awareness for me.

Women in tech at Intel

VentureBeat: According to Intel’s 2020 workforce data, the company demonstrated progress in the advancement of women to senior positions. What do you see Intel doing, from your position inside the company, to advance women into more senior positions?

From my vantage point, I see Intel doing multiple things. One example is that we have an Intel-wide senior technical women’s conference that is held annually. It started as a tiny forum years ago, where you could fit the attendees in a large conference room. The last time we got together, pre-COVID, we filled an auditorium.

This forum has been great to bring women technologists together from all geographies and diverse technical backgrounds, to make sure they can network with each other and with a broad range of senior technical leaders at the company. Also, this forum creates a venue for up-and-coming senior technical women to present their work, which gives them wider visibility and with that better opportunities for career growth.Intel also has mentorship and sponsorship programs.

I think the sponsorships are great. Leadership not only supports the program, but helps with pairing— matching technical women with leaders across the company so they get the opportunities. This is helpful because a lot of the time, I feel that many women don’t make networking a priority and recognize its value.

Above: A manufacturing technician in Kulim, Malaysia displays a tray of Ice Lake-SP-based 3rd Gen Intel Xeon Scalable processors.

Image Credit: Intel

VentureBeat: In the same workforce data report, Intel says it saw a decline in U.S. female representation. What are your thoughts on why that was the case?

Well, 2020 was quite unique, right? Anyone who was able to work from home did, which was very different and difficult for many. The boundary between work and personal lives blurred, and many struggled with it. Certainly, in my work experience, I saw that men or women who were helping younger children or other family members at home experienced additional stress. I think it’s no surprise that women struggled to stay in the workforce, not only at Intel but everywhere.

VentureBeat: Understanding that many women are working harder than ever to juggle the demands of their personal and professional lives, do you have any experiences or guidance to share?

Prioritize. Think what is important to you? Personal prioritization comes first. After that, I think there is a need for clear communication. Once you talk to your manager, your coworkers, and those at home, and you know what you can do, everything gets better. And when you’re happy, it will show in your work.

Don’t be afraid to reach out, either. After doing the due diligence and putting in hard work on a given challenge, don’t wait too long to ask if you think you need help to solve a problem. If you think you asked for help but nothing happened, I would encourage you to think through if you were clear enough. Did you list the exact help you needed? Did you say what would be the consequence of not getting the help? Were you clear about the tradeoffs that would need to be made? If not, do that first. Instead of bringing problems, bring solutions and alternatives to your leadership’s attention.

Once you clearly communicate a request for help, the managers and technical leads around you should be able to help you find a solution. If they cannot, they will be happy to guide you further. In my experience, disappointments are most of the time due to unclear communications.

I often ask my mentees if they have a role model. If they do, what are the qualities of that person that they really like? Are there any they dislike? I found for myself that the characteristics in a person you look up to will change as you grow in your career. Understanding your next set of goals with clarity will put you more in control of your career.

Ice Lake: Composing, then conducting

VentureBeat: You recently served as the lead architect for Intel’s Ice Lake-SP system-on-chip. Today’s data center-oriented processor architectures are massively complex, involving the very latest in hardware, software, and process. With so many pieces in play, what does a lead architect’s average day consist of to keep a handle on something so sophisticated?

As the lead architect, you wear many hats. Right at the beginning, during the initial phase, you are like a composer. You have a somewhat blank sheet in front of you. You have high-level requirements and targets in mind, and you architect to them.

After you create the design, you then become the conductor. You need to know each instrument, but you don’t need to be a master performer of each. And you need to think critically about how each instrument is used and coordinate to make sure everybody’s playing with the right intensity, color, transitions and harmonies. Whether an orchestra or an engineering team, that’s where the greatness comes from. You become the glue between all the IPs, accelerators, and cores in the SoC.

Later during execution, you are still the conductor, refining the performance of each instrument, solving new arising issues, culminating in the optimal complete performance.

VentureBeat: That’s a good analogy. As a trumpet player, I understand the challenge of controlling intensity. Most conductors also have instruments they’re trained on. In that regard, is there any aspect of the Ice Lake SoC that you had more technical influence on than the others?

My background is SoC performance, so scalability was my key goal. We added PCIe Gen4, we added more memory channels and controllers, and we added more cores. In the end, I wanted to make sure each SKU could hit the highest performance levels and maximize efficiency. So, I would say my biggest role was solving those SoC bottlenecks, making sure the coherence flows are optimized, and when the whole processor is active, we have a fair distribution of the limited resources. 

VentureBeat: With Ice Lake-SP now shipping in production, were there any learnings you’d like to share as a leader on that project?

Definitely. In every program, we learn new things. The use cases for more sophisticated SoCs are evolving and changing. Now we have a greater focus on AI and deep learning. At the same time, there is an increased need for general-purpose computing. You also have higher security, virtualization and networking demands.

So, at the end of the day, for new technologies we’re putting into our SoCs, we need to be proactive, building the test vehicles and software enablement ahead of the hardware. Sometimes this happens head-to-head. Sometimes the software isn’t available when the hardware is ready. And those are the times when we take longer to validate. I’d say we’re always working to improve the pre-silicon validation capabilities for software and hardware together.

VentureBeat: What’s  your next project?

My transition to the next program actually happened gradually during the last phases of Ice Lake’s execution. Today, I am 100 percent focused on the next-generation Sapphire Rapids program, and I’m looking at the performance architecture of that program.

In general, when a program ends, we do postmortems and try to identify learnings that we communicate to our architecture and design communities. After that, some folks move on to the next CPU architecture. I was wearing the performance architecture hat for Ice Lake; I will continue to focus on performance architecture for Sapphire Rapids.  I can’t wait to talk with everyone about the innovative features we’ll deliver in Sapphire Rapids – oh, and of course, the amazing performance.

The service providers may be storing the organization’s data with a cloud provider, and that could expose the organization to risk if the storage is not properly configured. For example, a service provider could host a website within a cloud storage bucket, such as Amazon Web Services Simple Storage Service (S3). The default configurations can lead to security issues because they make the server object and file contents publicly accessible.

Panorays assessed the external cloud infrastructure of third parties’ cloud providers as part of its research. More than 20% of companies have their cloud storage in a single region. This is not surprising, since it’s cheaper and easier for companies to deploy single-region architecture. However, this practice can be problematic when faced with disaster recovery, which is why it’s not recommended.

Panorays checked if specific cloud services were exposed to the public because exposed cloud infrastructure resources increases risk and the organization’s attack surface. From the tens of thousands of storage buckets found in our assessment, 5% were found to have public browsing/listing permissions.

Panorays also checked if the services were located in a single geographic region since having different regions is recommended for business continuity.

These buckets belonged to professional companies and not private individuals, which may explain why the figures were lower than expected, said Demi Ben-Ari, Panorays co-founder and CTO. “Nevertheless, we expected this to be even lower, considering that open buckets remain a pressing problem that has gained a great deal of publicity lately.”

Read the full research from Panorays.

Newzoo is still predicting the game industry will shrink in revenues about 1% to $175 billion for the first time in years in 2021. But the company said the COVID-19 pandemic has accelerated gaming trends already in motion, leading to an expanded gaming audience and through-the-roof engagement for much of the past year.

Will engagement continue after the pandemic?

Newzoo believes people enjoying renewed freedom (and spending more disposable income on non-gaming activities) will spend time away from games, but it also believes that the trend of using games as social spaces is here to stay.

The research focused mainly on gamers in North America and Western Europe, where gaming replaced in-person socializing for players. Fifty-one percent of gamers in Western Europe and North America reported spending more time playing games since COVID-19 began.

Newzoo’s data also shows a clear connection between age and increased time spend playing games. As you can see in the image below, the younger the age group, the higher the likelihood of playing.

Above: Newzoo says younger gamers like social and competitive games.

Image Credit: Newzoo

Asked why they played, a third of 10-20-year-olds cited socializing as a major reason for playing. The group is also more likely to cite competing and improving skills as major factors. Meanwhile,
older age groups are likelier to value relaxing experiences.

So it’s unsurprising that online multiplayer modes are more popular among younger gamers across the two regions. Almost half of gamers aged between 10 and 35 claimed they play online multiplayer modes most vs. 36% of those aged between 36 and 50.

Game-related engagement goes beyond playing

Newzoo noted that watching games also became more popular. By the end of this year, the games live-streaming and esports audiences will hit 728.8 million and 474.0 million, respectively.

But gaming fandom is much more than just playing games, viewing live streams, and owning peripherals. Not only do many gamers play titles and watch their favorite creators live on sites like Twitch and YouTube; many also read, watch, and listen to industry news and developments via websites, apps, and podcasts.

The also actively contribute to discussions on Reddit, Discord, ResetEra, and other communities. And they discuss developments with friends, family, and peers (virtually and in person). This goes even for casual gamers.

In the past six months, 80% of gamers in North America and Western Europe talked about games with peers. Sixty-one percent engaged with gaming communities. And 60% visited gaming  websites/blogs/listened to podcasts.

Community gamers

Above: Newzoo’s report on gamer behavior talks about community gamers.

Image Credit: Newzoo

Based on 33 markets, 10% of all game enthusiasts are what Newzoo calls “community gamers,” and social engagement and viewing are more important to this persona than playing. Community gamers also skew male, have an average age of around 29, and are likely to enjoy value non-competitive elements of games more, such as strong narratives and stories, exploration and open worlds, and specific themes.

The unique ways community gamers engage with games also means the ways brands and publishers can reach them are also unique. This uniqueness may be important, as this group drives online conversations by nature, making community gamers inherently influential on the online narrative around games, game-related brand promotions, and more.

This group, and many others, has been engaging with games even more throughout the pandemic, and Newzoo believes that the trend of using games as social spaces will stick. Gaming was always massive, but now it’s even bigger. It’s bigger than movies, and is a true entertainment innovator, Newzoo said. Thanks to the burgeoning metaverse trend, social aspects of games will play an even more vital role going forward, and the community is at the heart of it all, Newzoo concluded.

The goal of the Dremio Dart Initiative is to increase the performance of those SQL queries by a factor of five in the next 12 months, said Tomer Shiran, company founder and chief product officer at Dremio.

Rather than copying data into a proprietary data warehouse deployed in the cloud or in an on-premises environment, Shiran said it’s more cost effective to create a data lake using its SQL engine to directly access data stored, for example, in an Amazon S3 and Azure Data Lake Storage (ADLS) service. That approach also eliminates the need to aggregate tables, extract data, or employ a separate online analytic processing (OLAP) cube to structure data in a way that is compatible with SQL, Shiran added.

Dremio is now allowing IT teams to deploy unlimited table sizes spanning an unlimited number of partitions and files, with all data becoming near-instantaneously available as it is added to the data lake. Dremio is also adding support for manifest-based metadata and version management to manage those larger data sets.

In general, Dremio is trying to make it simpler for end users that have SQL expertise to launch queries directly against a data lake or enable of the users of business intelligence (BI) applications to launch queries via the Dremio engine without having to know SQL. Regardless of the approach, the goal is to make sure organizations are not locked into a proprietary data warehouse deployed in a cloud in the same way they are in an on-premises IT environment. “They are not in love with that being their future again,” Shiran said.

Optimizing queries for the data lake

As part of the Dart Initiative, Dremio is optimizing query planning by automatically gathering statistics about the underlying data, which enables the Dremio query optimizer to determine the optimal execution path for any given query before data is loaded.

Dremio is also adding support for query plan caching, which eliminates both overhead and latency for repeated queries, in addition to a high-performance compiler that enables much larger and more complex SQL statements while employing machine learning algorithms to reduce the amount of compute resources required to launch SQL queries. Cloud storage read operations make up 30% to 60% of query execution costs in some workloads, Dremio says, and the company is reducing the amount of data read from cloud object storage by enhancing the scan filter pushdown capabilities it provides.

The company is also broadening its support for SQL to include support for additional functions, operators, and SQL grammar constructs. The Dart Initiative provides a significant boost in the performance of end-user queries with complex expressions by extending the capabilities of Gandiva, a toolkit that enables vectorized execution on modern processors within the in-memory buffers in Apache Arrow, an open source columnar data format Dremio co-created.

Finally, Dremio has enhanced its ability to support the orchestrated refresh of hundreds of reflections employed by BI applications within multi-tenant environments. Future Dart Initiative phases will advance acceleration management through improved refresh granularity, consistency across related reflections, and improved refresh monitoring and restartability.

Those capabilities complement technologies such as the open source Apache Iceberg table format, which Dremio employs to enable multiple instances of its engines to work together, and Project Nessie, which brings Git-based semantics and workflows to data lakes, Shiran said.

It may be a while before organizations fully appreciate the degree to which open data lakes might obviate the need for proprietary data warehouses. The challenge is distinguishing between what truly constitutes a data lake versus just another data warehouse that happens to be deployed in a cloud.

In recent years, large-scale AI language models have shown promise in generalizing to tasks including writing code, implying that humans’ work may be one day supplemented by AI systems. But while some studies show that language models can translate code and fix compilation issues, there’s been little work on rigorously testing the coding ability of models given general coding problems.

That’s why a team of researchers at the University of California at Berkeley, Cornell, the University of Chicago, and the University of Illinois at Urbana-Champaign created APPS, a benchmark for code generation from natural language specifications. Unlike prior work on code generation, which mostly focuses on code translation and pseudocode-to-code, the researchers tested models on their ability to take specifications and write code that meets these specifications.

Their work comes on the heels of the release of IBM’s Project CodeNet, one of the largest open source dataset for benchmarking around AI for code. But CodeNet centers around the problems of code translation, code similarity, and code constraints. APPS is broader in scope, evaluating models not only on their ability to understand coding syntax but on their ability to comprehend task descriptions and create algorithms to solve these tasks.

“APPS enables robust evaluation of models along several dimensions, providing a precise and comprehensive view of code generation ability,” the coauthors wrote in a paper detailing their work. “If a model were to perform well on APPS, this would indicate an ability to flexibly use data structures and programming techniques, as well as an ability to correctly interpret diverse task specifications, follow instructions, and understand human intent.”

APPS contains 10,000 programming problems in Python, Java, and C++ ranging in difficulty from introductory to coding competition challenges, as well as a bank of over 130,000 test cases and more than 230,000 human-written solutions for evaluation. The test cases were chosen to create a gold-standard metric for model performance, including correct functionality across edge cases. And most were taken from open access coding websites including Codeforces and Kattis.

The introductory problems in APPS, which include counting the number of appearances of a substring and finding if a string is a palindrome, can be solved by programmers with 1-2 years of experience without requiring algorithms. The intermediate, interview-level problems are more difficult in nature and at the level of questions asked in typical technical interviews. As for the competition-level problems, they’re even more challenging and representative of those in high school and collegiate programming competitions like the United States of America Computing Olympiad (USACO).

Results

The researchers tested several types of models on APPS, including OpenAI’s GPT-2, GPT-3, and an open source version of GPT-3 called GPT-Neo. In experiments, they discovered that the models could learn to generate code that solves easier problems but not without syntax errors. Approximately 59% of GPT-3’s solutions for introductory problems had errors, while GPT-Neo averaged 3%. Moreover, the best-performing model — GPT-Neo — attained only 10.15% accuracy (excluding edge cases) and 1.12% strict accuracy (including edge cases) across introductory-, interview-, and competitive-level problems, indicating that there’s substantial room for improvement.

“These results position code generation as a challenging but tractable testbed for large-scale language models … Writing code to meet specifications in natural language is an economically valuable task with widespread social implications should it be solved, in that it could eventually facilitate malicious code generation and one day result in job automation. As large-scale language models have the potential to make significant progress on code generation, it is essential that we begin to track advancements on this task,” the researchers wrote.

Several efforts are underway to create viable AI-powered coding tools, including Intel’s ControlFlag, which can autonomously detect errors in code. Codota is developing a platform that suggests and autocompletes scripts in Python, C, HTML, Java, Scala, Kotlin, and JavaScript. Ponicode taps AI to check the accuracy of code, and DeepCode offers a machine learning-powered system for whole-app code reviews (as does Amazon). Perhaps one of the most impressive projects to date is TransCoder, an AI system Facebook researchers developed that converts code from one programming language into another. Another contender is a model from OpenAI that was trained on GitHub repositories to generate entire functions from English-language comments.

According to a study from the University of Cambridge’s Judge Business School, programmers spend 50.1% of their work time not programming; half of the rest of their time is spent debugging. And the total estimated cost of debugging is $312 billion per year. AI-powered code suggestion and review tools, then, promise to cut development costs substantially while enabling coders to focus on more creative, less repetitive tasks.

The Santa Monica, California-based Mobalytics rebuilt all its core features with an “extreme” focus on quality, speed, and performance, said cofounder Amine Issa in an interview with GamesBeat. That’s pretty gutsy, as millions of players use the tool.

Issa was obsessed with high-performing athletes, and he studied them as only a statistician would. He extended that study to esports players. In 2016, Issa cofounded Mobalytics with Bogdan Suchyk and Nikolay Lobanov to build the “Moneyball of esports.” They participated in that year’s TechCrunch Disrupt event and won the Startup Battlefield contest.

Creating the GPI

The company created its Gamer Performance Index (GPI) platform to measure a player’s performance in League matches. It analyzes the data and gives the player insights and recommendations for how to improve. And it provides that info in an easy-to-understand dashboard. The system uses in-game data for players that is available through publisher Riot Games’ applications programming interface for League. Operating with Riot Games’ application programming interface for data, Mobalytics was able to raise $11.25 million last year.

Now the company has refined its user interface so that the information is all in one place: champions, profiles, live companion, and more, without switching between tools. With the newly updated product, Mobalytics focused on convenience. It delivers info to you automatically before, during, and after the game.

It has an upgraded Live Companion feature with support for the popular All Random All Mid (ARAM) player-vs.-player mode in League of Legends. The Live Companion has improved speed and stability, an OP in-game overlay with helpful insights and notifications.

Above: Mobalytics cofounders Amine Issa (left) and Bogdan Suchyk.

Image Credit: Mobalytics

“There was always a standard for measuring performance, right? You have all the numbers like K/DA (kills, deaths, assists),” Issa said. “There are a bunch of different stats that people look at but they don’t necessarily provide tremendous amounts of meaning and precision.”

He added, “There are a lot of different ways to track the performance of players. So our goal was to create the Gamer Performance Index (GPI) and make it a standard for understanding player performance. We built out the platform four years ago, and we had a bunch of players playing League of Legends and we used that data. And over time, we kept iterating and improving it. And we started to work with Riot Games. And in the last year was we became the visual data partner.”

As it was creating its GPI, the company had to pull back and think more about everyday players, rather than the highest-performing esports athletes, to design a product that helps everyone.

“Not all the people are willing to focus and spend too much time on improving,” said Suchyk. “They just want to win their next game. For the last two years, we have been developing our product to transform it into an ultimate gaming companion that uses a lot of data in the background. But instead of giving you the data to digest, we’re trying to use this data to recommend specific small things that you can use during your next game.”

New and Improved GPI

Above: Mobalytics’ newest Live Companion screen.

Image Credit: Mobalytics

Players can also see a cleaner way to assess their skills and see their progress. In general, it’ll be much easier to find out strengths and weaknesses, and track the ups and downs of individual skills. Progress Tracking is now within the GPI.

Mobalytics said it wants to assure players that challenges will only help them win more and improve skills one game at a time. It will also get rid of some features like dashboard, progress analysis, and player comparison. In the coming months, they will either transform or be replaced by something bigger and better in the future.

Players get additional information available while they’re playing and then afterward assess their overall performance. It also helps players understand their performance during the long term. It will suggest which characters or roles you need to play better, or practice more.

Finding the best amateurs

Above: Mobalytics wants gamers to fulfill their dreams.

Image Credit: Mobalytics

Mobalytics also helped Riot Games’ League Championship Series to develop a new amateur ecosystem with the hope of closing the gap between amateurs and professional players. The hope is to use the data tools to help uncover new talent for esports.

As for finding new talent, no single stat stands out as the best. But the games are complex and one player who is good at one role may not be good at another.

“We try to look at the data more holistically,” Issa said. “Like how you do at a particular role and what the data says about your level of aggression.”

The company will focus on the player who has the best stats for a particular role and highlight that. The GPI makes use of eye-tracking metrics, as the best League of Legends look at parts of the screen very efficiently to absorb information and figure out what to do next.

“We use eye-tracking to tell their map awareness and how well they’re making decisions around the information they’re receiving on the map,” Issa said. “And we index that. And that’s been especially useful to pro players. The problem with something like that is that for us to heavily index on it, as it doesn’t offer that much benefit to our mainstream players because not many people own eye trackers yet.”

Issa said, “League is a very complicated game. Stats are highly contextual, depending on the role and the team’s actual performance during a season.”

Expanding to new games

Above: Mobalytics provides a Live Companion for League of Legends.

Image Credit: Mobalytics

During the past year, Mobalytics branched out and added overlays to help players with Valorant, Legends of Runeterra, and Team Fight Tactics. Issa wants to expand to more games, but the latest update for the League of Legends companion was important to do to satisfy the current users.

The company has about a dozen people in Santa Monica, but many of its staff are distributed around the world in places such as Australia, Ukraine, and Russia. All told, it has less than 50 people.

With data on four games now, the company plans to expand to other games in the future. The pattern among most of the great games is that they’re very easy to get into but the mastery curve is long and deep, said Suchyk.

“Providing a tool that acts as a companion and helps them is a win-win for everyone who wants to enjoy a game but doesn’t necessarily have the time to master that game,” said Suchyk.

Riot Games understands the importance of the data because it is sharing it with companies like Mobalytics. It can predict who might win a match, but it isn’t focused on predictive technology for gambling purposes.

Mobalytics has its own challenges finding what it calls game experts. These are people who are really good at the game and can thoughtfully articulate their expertise in a game and feed that information to Mobalytics so that it can sift the data and provide insights. It also uses machine learning and data scientists to figure out how to access the relevant data and provide that to players as quickly as possible.

“It’s very hard for machine learning to understand a new champion and rate its strength,” Issa said.

The bottom line: Good decision-making is what wins all games, Issa said.