“What’s the state of the global chatbot economy today?”

That’s a question I’ve been asked a lot lately and one that I’m hoping to get more answers to by attending the second edition of Chatbot Summit in Berlin at the end of the month.  Since bot mania took over the tech world nearly a year and a half ago, not a single month has gone by without significant news coming from startups and the usual suspects (i.e. Facebook, Google, Amazon, Microsoft, etc.). The space has been hyperactive since the beginning, and it doesn’t show any sign of slowing down.

That said, a lot of brand managers and chief digital officers are still new to the bot world, so I thought it would be handy to go back in time in order to understand where the bot ecosystem stands today.

By the numbers

As of mid-2016, more than 11,000 Facebook Messenger bots and 20,000 Kik bots had been launched. Over the last year, you could’ve shopped New York Fashion Week pieces from the Burberry bot, asked the Starbuck’s Pumpkin Spice Latte bot its favorite book, or sent emojis to the British Airways bot to get vacation recommendations. Fast forward to this stat from April 2017: 100,000 bots were created for Messenger alone in the first year of the Messenger platform.

VCs have started paying attention to chatbots too. In the first six months of 2016 alone, $58 million was invested in chatbots and 29 new bot startups were founded. Slack launched an $80 million fund last year to invest specifically in bots running on its platform. In fact, there is even a website that tracks all the chatbot financing as it happens.

Race between the tech giants

For the past few years, industry giants have been competing to build the best chatbot and the best bot platform.

“Bots are the new apps,” Microsoft CEO Satya Nadella said in March 2016, arguably setting off the chatbot revolution. But the company took to the movement with a bit of a rocky start last year. Tay, an AI chatbot made for Twitter, got taken down about 24 hours after its launch when its tweets became racist and hate-filled. After that, Microsoft launched Zo, a bot for Kik , in December 2016 and then Ruuh, an AI bot for the Indian market, this March.

Facebook’s exclusive AI butler bot, named M, was released in 2015 to an exclusive group of 10,000 users. It’s one of the few bots that aren’t rules-based; instead, this virtual assistant is built on algorithms and, when faced with complex requests that those algorithms can’t handle, powered by humans.

Google, on the other hand, has been busy amassing tools to support chatbots. For example, the company acquired API.ai, a conversational user interface platform used by over 60,000 developers, for an undisclosed amount last September. And in May 2017, Google debuted a chatbot analytics platform called Chatbase that can make suggestions about how to improve bots based on user analytics.

In April, Amazon opened Amazon Lex to the public. Lex is a platform that gives developers access to the same tools that power its digital assistant Alexa, and it bundles speech recognition, text recognition, and conversational interfaces. In the same week, Amazon also announced its AWS Chatbot Challenge — a competition to create a lifelike user experience.

Samsung bought Viv, billed as a more powerful Siri, last October for more than $200 million and in February launched Nexshop Training, a virtual assistant chatbot built for training retail personnel.

Apple’s response to the chatbot movement, in a very Apple-like move, didn’t involve any chatbots. With iMessage Apps, developers can build app extensions that live inside an “app drawer” in iMessage. For instance, in a Square Cash iMessage app, there would be a green widget instead of a keyboard so you can input the amount of money to deposit.

Brands are joining the conversation

Following the leadership of tech giants, brands are getting in on the fun. Last December, 80 percent of brands said in a survey that they were using or plan to implement a bot by 2020.

Hundreds of brands have created their own bots, from Casper’s text-based Insomnobot, which will stay up with you during your sleep-deprived nights, to Sephora’s beauty bot that will recommend makeup tutorial videos via Kik and Messenger. You can also find news bots that update you daily, sports bots that tell you the score the second the game ends, and finance bots to check your account balance or send money.

The next frontier

All this has happened in the last year or so. Imagine where we’ll stand in 2018, as interest in bots extends beyond Silicon Valley and continues to spread across Madison Avenue, where ad agencies are increasingly extolling the virtues of chatbots.

Chatbots themselves are getting better and more ubiquitous, and the tools for building and analyzing them are becoming more plentiful. So what’s next?

First, bots will need to be interoperable across platforms. You should be able to find the same bots on Facebook, Kik, Alexa, Google Home — any messaging platform you choose — and they should all be able to remember your data across platforms to create a seamless user experience.

Second, we’ll need to find ways to make chatbots more easily accessible. Discoverability is still a major problem in the bot ecosystem and one that needs to be addressed in order for chatbots to become mainstream.

One thing is sure: The technology is expanding exponentially. Personally, I cannot wait to see where we’ll be in 12 months. Probably light-years from where we are now…but either way, this is just the beginning of the chatbot revolution.

Étienne Mérineau is the cofounder and head of conversation design at Heyday.ai, a chatbot development platform for brands.