(Editor’s note: this is a guest post from Ha Duong, a former associate at Techstars, product marketing professional, blogger and an aspiring entrepreneur from Germany.)
We have seen a fast growing interest in current activities around AI startups and research in the last couple of months.
Headlines like “2016 was the year AI came of age”, “AI was everywhere in 2016”, and “The Great A.I. Awakening” were all over the media in the ending weeks of 2016 and we are curious about what 2017 will bring.
I found particularly interesting that the current applications, future potential, and possible risks even attracted interest beyond the tech community through TV shows like Westworld, coverage on traditional media and even Obama’s farewell address.
Sadly, for many of us tech enthusiasts here in Europe, we sometimes feel like there is way less movement on this side of the Atlantic than in the Silicon Valley. However, with major acquisitions like DeepMind, Magic Pony Technology, Movidius, Vision Factory, and Dark Blue Labs, Europe has shown that it is actually leading the way in AI and machine learning. I wanted to show the great potential of European AI startups and eventually put together a list that I would like to share with all of you.
> You can access the “European AI Startups“ Google Sheet here.
I’ve been looking into the European AI and ML landscape for a while and I realized during my research that this list could be valuable to many of us, especially if you are:
- an investor looking into the European AI / ML space
- an entrepreneur doing market research about potential opportunities in the landscape
- a ML startup selling to other ML companies that want to fill up the lead list
- a student looking for the next rocket ship to jump on board
- a researcher looking for use cases to apply their newest ML models to
- a tech enthusiast who wants to see what is happening out there in the AI / ML space
Many others have put together great overviews of AI or ML startups across the world like MMC Ventures’ UK AI Landscape, the European Machine Intelligence Landscape from Project Juno, and The Current State of Machine Intelligence 3.0 from Bloomberg Beta’s Shivon Zilis.
My list of European AI startups will add to their amazing effort by giving granular data about the individual startups to everyone.
I included in my list all private companies with a presence in Europe which do something around AI applications. Thereby, AI — the science of intelligent programs — is a very broad field and goes way beyond what is known as machine learning. Frank Chen from Andreessen Horowitz listed the subfields of AI as: 1) reasoning, 2) knowledge representation, 3) planning and navigation, 4) natural language processing, 5) perception, and 6) general intelligence (including emotional intelligence, creativity, moral reasoning, intuition, etc.).
Machine Learning is a subset of AI that goes beyond rules-based programs and solves problems through sophisticated mathematical and statistical optimization methods. Recent AI breakthroughs were often achieved with Deep Learning — a subset of Machine Learning that involves neural networks with multiple layers. Especially, Deep Reinforcement Learning and (Deep) Unsupervised Learning will be the next hot topics around AI research and applications.
Going through the list of 600+ companies, I realized that only a few startups focus on AI fundamentals with a strong research character while many more companies see the opportunity in Applied AI. I didn’t emphasize in my list whether a startup is working on proprietary AI technology or only applying 3rd party APIs and services in its applications.
Typical application areas can be categorized by industry (finance, insurance, manufacturing, healthcare,…), function (marketing, sales, developer tools, HR, compliance,…), or technology focus (computer vision, natural language processing, VR/AR, autonomous systems,…).
Here is a short summary of my findings:
- 75% of the European AI startups were founded in 2013 or later
- About 1/3 of the companies raised a funding round in 2016 or later
- Of those who raised since 2016, seed investments have an average round size of $1.6M and a median size of $1.1M, while Series A investments have an average round size of $8.8M and a median size of $6M
- VC funds with notable investments in the AI space or a data-driven investment thesis are Playfair Capital, LocalGlobe, Balderton, Octopus Ventures, Index, Northzone, Episode1, Sunstone and BlueYard
- Common acquirers are tech giants like Google, Apple, Twitter and Intel but more traditional companies are entering the field
- About half of the companies have their headquarters or a strong presence in London where the ecosystem and funding capabilities are still stronger
- Other strong hubs with presence of AI companies are Berlin (13%), Paris (8%), Amsterdam (3%), Zurich (3%), Dublin (2%), and Stockholm (2%)
- Industry focus: Finance: 7%, Security: 5%, Healthcare: 4%, Manufacturing: 1%
- Technology focus: Computer Vision: 13%, Natural Language Processing: 9%, Virtual/Augmented Reality: 2%
- In terms of domain name, .io (11%) is still more popular than .ai (5%)
This categorization and the resulting analysis are obviously highly subjective and shouldn’t be taken as a perfect representation of the market. However, it should be able to show some certain trends. Most startups enter the market with a niche solution focusing on their core strength in one of the machine learning fields.
Some companies like Seldon, Dataiku or understand.ai see themselves as enablers for AI and offer platform or service solutions that empower other AI applications. Others like Oxbotica or Daedalean go big from the start and approach their visionary goals to change the world. All of them have an immense impact on the local tech ecosystems and they will hopefully continue doing so for the future.
I hope my list of European AI startups can help you get an overview of the market and detect new exciting business models you have not heard about. If there is some data you can add or correct please do so and feel free to shoot me a message with thoughts or feedback.
Featured image credit: jcwait / Adobe Stock