Rudina Seseri, founder and managing partner of Glasswing Ventures of Boston, has over 14 years of experience in technology investments in startup companies in a range of fields including digital media, marketing technologies and robotics
She has served as a lead investor and director on the boards of companies including: Celtra, providing creative content management for digital platforms; CrowdTwist, a SaaS-based multi-channel loyalty and analytics platform; SocialFlow, for social media optimization; Statisfy, providing writers, bloggers and content creators with a set of engagement tools; and Jibo, a social robot for the home.
Rudina will be sharing her knowledge as a speaker at the AI World 2017 Conference and Expo; we spoke this week about how she views artificial intelligence in the context of her investment strategies.
How would you describe the pace of AI adoption?
My partners and I have been in venture capital for 11 years. We have seen AI grow dramatically in the past 6-7 years. If you look at our portfolio of companies and the maturity of AI, you will see parallel lines. But it is still very nascent.
We are seeing AI becoming a layer across the enterprise. Our fundamental view is that AI is a layer for any enterprise solution going forward. Some markets are lower hanging fruit. Over time, all markets and verticals will leverage AI. Eventually from a consumer and healthcare perspective, there will be nothing that doesn’t have AI in it.
Are there companies you are interested in we may not have heard about yet?
I can describe some without giving you the names just yet. We have a lot of interest in new sources of data. If you think about AI, the various facets of learning from supervised, unsupervised, deep learning – one could argue that what has a lot more value than the network is really the data. So we are looking at some next generation plays for companies to become the hubs for data for the next generation of AI.
We are also looking at leveraging learning and prediction in cyber-security to figure out the links between data and security controls, such as when data should leave the enterprise. Other themes are around new platforms and new interfaces for B2B for consumers, and how they engage. We are researching new ecommerce engines to enable new consumer experiences.
And we are looking at new business networks, what next-generation LinkedIn-type platforms look like. We spend a lot of time looking at the recruitment market.
What’s happening with AI and computer vision?
The question there is from a venture perspective, like Nuance Communications did speech and imaging, will there be one play for vision? Will you have a horizontal platform that can be applied to different use cases, or will it stay within vertically-integrated solutions? So will vision be proprietary and integrated vertically by different suppliers, or will the vision component be a part of a platform play. I’m going back and forth on that.
How is the role of data scientists evolving?
True data scientists are a very, very, very scarce resource. The genesis of the new AI wave goes back to deep learning, back to 2010 to 2012. As a field, the new classes of graduates do not have a lot of depth. They are just now ramping up. The few you can find are getting very expensive. So there is a race to get those guys.
We have an advantage on the east coast because a lot of them are coming from Carnegie Mellon University, Cornell, NYU, Columbia, and even Harvard now. So we have a concentration of the relevant universities on the east coast. The Facebook AI team is headquartered in New York. The machine learning people for Uber are in Pennsylvania. So we have a concentration of talent, but we need even more of it.
Then you see companies slapping the data scientist label on someone who studied math and statistics. It makes a huge difference if you have the real experts or not.
Why is Boston a center for cyber-security?
I would recast it as we have pockets on the eastern corridor that have an advantage because the talent is there. You have a critical mass of talent. Cyber-security as a market has done well on the east coast.
How is robotics coming along in your view?
I am one of the early investors in Jibo. I have a huge interest in intelligent devices, the move to connected devices. The role of robotics and data are tied very closely to each other, and will be at play in the relationship we have with devices in the form of robots. They will be helpful in our lives. We have not even scratched the surface.
What would you like to be reading about in AI Trends?
You should take a look at data-heavy businesses. They are either disrupting themselves or someone is disrupting them. Hedge funds, large banks, all have huge internal projects, whether we are hearing about them or not, around AI. And a lot is going on in healthcare.
Brands in retail have gone from offline to digital to mobile and now, tying all the data together, they have entered the world of multi-channel. So now whether you are online, in the store, on your mobile device, accessing Facebook, all these tools have ties via multi-channel access. Their internal effort with AI is focused on learning about the customers, so that the next time you send a coupon, you know exactly what the customer is looking for.
Anything to add?
In my view, the speed of adoption of AI as a layer is going to occur doubly faster than anything we have seen in the recent past. The acceleration and adoption will be incredibly transformative. Like with Talla, the path to automation will happen and it will happen fast.
By John P. Desmond, AI Trends editor
This content is original to AI Trends.