Check out the on-demand sessions from the Low-Code/No-Code Summit to learn how to successfully innovate and achieve efficiency by upskilling and scaling citizen developers. Watch now.
Artificial Intelligence has become a buzzword for investors of late, many of whom recognize its enormous potential to become the most game-changing technology since the industrial revolution. Indeed, the projected impact of AI is likely to be greater than all prior tech trends combined, and savvy investors would be wise not to miss out.
From an investor’s point of view, you can divide the AI sector into a few major sub-sectors: infrastructure, algorithms, platforms, and applications. The infrastructure segment includes technologies and companies that provide the underpinnings enabling AI: machine learning, deep learning, natural language processing, and computer vision, including cloud infrastructure, specialized semiconductors, large-volume storage devices, low-latency databases, edge-based computing elements, and more.
On the algorithm side, one would primarily count neural nets, classification, and clustering algorithms, good old Bayesian networks, and hidden Markov models.
AI platforms implement algorithm families on proprietary or standard infrastructure, allowing rapid development of applications as well as standardization of business processes and greater transparency, reproducibility, and collaborative efforts.
Intelligent Security Summit
Learn the critical role of AI & ML in cybersecurity and industry specific case studies on December 8. Register for your free pass today.
Finally, the applications sector focuses on the use of AI components and methods in a variety of different use cases, solving and improving operations for the enterprise, vertical industries (e.g., automotive, agriculture, finance), and consumers.
In short, AI is used today primarily to improve on tasks that people don’t do well, addressing challenges such as speed, latency, repetitive tasks, or accuracy levels. Humans are very good at processing unstructured data. However, humans cannot process massive amounts of structured data very well. For example, with regards to computer vision, AI improves or enables a wide variety of applications including pick-and-place robotic arms, health care diagnostics, autonomous cars, and more. Applications in natural language processing include smart home voice interfaces, text analytics, and bots.
How will this play out in our daily lives? In the near future, we’ll be able to utilize AI to describe what’s happening in a foreign-language video, to search for documents related to deep learning and summarize them, to decide which image of a human eye displays symptoms of diabetic retinopathy, and even to bring us a cup of tea from the kitchen.
This mind-boggling potential led to an explosion in global investment in AI over the past five years, reaching $15 billion across 2,250 deals, with more than 200 mergers and acquisitions. Many AI investments are made in specific vertical applications such as health care, commerce, fintech, cybersecurity, and sales and marketing.
The venture capital firm where I am general partner, JVP, has invested in AI technologies over the past decade, with 12 current companies in our portfolio applying AI to fields such as cybersecurity, fintech, industrial IOT, IT dev ops, and enterprise software, including such firms as ThetaRay, Loom Systems, and ScadaFence. Several other portfolio companies are building AI-enabling platforms such as Iguazio and Upsolver.
We believe startup opportunities in AI can be found in 2018 in various clusters, including:
- Platforms and technologies that enable the use of AI, streamline AI modeling, and combine it with traditional software development.
- Algorithmic approaches that implement unsupervised learning, transfer learning, and reinforcement learning for specific verticals. This will bring AI closer to the human brain’s ability to transfer knowledge, learned patterns, and behavior between different domains and explain the reasoning behind such decisions.
In a world where differentiated data sets are the drivers of growth, startups that are able to gain access to proprietary data will have a significant advantage. An example of how startups can achieve this is through simulators that emulate the real world and address the scarcity of data for training, or through new algorithmic approaches for specific problems that allow researchers to train machine learning with a much smaller set of data rather than the huge data sets they currently require.
That said, data-rich companies with access to proprietary and continuous data sets enjoy a competitive advantage and are well positioned to augment their businesses with AI applications in a differentiated manner. That means startups, while enjoying expertise, experience, and a deep understanding of the capabilities of AI, may nevertheless be at a disadvantage due to lack of data. Resource-rich giants such as Google, Facebook, and Amazon will likely win the race to procure scarce talent needed to apply AI approaches, which includes subject-matter experts, scientists, and AI engineers, by either poaching talent or offering early-stage acquihire exits for startups.
In short, enormous opportunities exist for startups and early-stage investors in the burgeoning world of AI, but both will need to act wisely to win the very real race towards artificial intelligence in the future.
Yoav Tzruya is general partner at Israel-based VC firm JVP.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.