AI Summit Thoughts
Last month, our company Artificial Solutions attended the San Francisco’s 6th edition of AI Summit, in yet another successful installment. The venue was quite beautiful with grandiose ballrooms, worthy of a renaissance film set. It was unusually hot for the city and the attendees were literally sweating, yet the atmosphere was still pulsing with zeal. We’ve participated in all three locations, London, New York, and San Francisco from inception. But nothing was as jam-packed as this time around.
The sponsors all had contributions to make. Microsoft displayed their smart sunglasses and the recently released Seeing AI app, that helps blind people see by describing, visuals captured through their phone. Intel dove into complementary learning, and the benefits of a more complete AI system, in which both deep leaning and cognitive memory work together. While Facebook reminded everyone of their widespread consumer coverage by addressing personalization in AI.
There were highly technical talks for the techies, and more general industry trends for the newbies. Even actual implementation of AI in the workplace was addressed. And while blockchains are favored by the financial sector, it is AI that can quickly be tested and adopted into the workforce with less overhead cost.
Personally, the most interesting comment came from Frank Chen of Andreessen Horowitz during his Q&A after dazzling everyone with his presentation, The Promise of AI. His speech was quite positive, exploring everything newsworthy from self driving cars to the applications of image recognition software.
The TensorFlow powered cucumber sorter by Makoto Koike was given as an example of a real-world AI application and even overlapped with the talk on machine learning given by Magnus Hyttsten, of Google: ‘A case of simplification & scale’. The speech gave the impression that AI was truly here to stay. However, when asked about the future of Chatbots, Frank was quick to point out that language is a very tough thing to crack. While optimistic of the adoption of machine learning and the likes, NLP requires much more iterations than one might expect. The consensus was that while we will get there, it might take more time than we have become accustom to expect.
This was satisfactory for me to hear because the popularity and confusion about chatbots has troubled me. Clearly the interest is there. All the big players have come on board to acknowledge that voice is the next big tech investment, if the sheer amount of NLP startups has not made it obvious already. However, conversational AI is not as easy as one might be led to believe. So many have overpromised and underdelivered, when the aim should be to set expectations.
Every Fortune 500 company that is looking to test natural language solutions, should be ready to work with innovation and truly invest into this initiative, while at the same time holding their customers’ hand. It is a balancing act, but it is extremely valuable in the end. If you want to get into specifics read our white paper 10 Things to Ask Your Next Digital Employee. Otherwise enjoy this surreal and thrilling AI journey. I’m eager to see where it goes.
Also check out our booth at NYC AI Summit in December.