On 25 June, Cosmo Tech and Braincities hosted the Limits of AI Conference, held at BpiFrance in Paris.
With the tagline the best way to prepare for the future is to understand it, the Limits of AI conference looked at the challenges and limitations that modern-day artificial intelligence poses to us today. The conference featured thought leaders and from various backgrounds all centered around the same question: what if the limits of AI were defined by the limits of our own humanity?
Giving the opening remarks of the day was Kazé A. Onguene, the CEO and founder of Braincities. Braincities is a startup that aims to provide context-aware recommendations for decision makers, by identifying and analyzing emerging trends. Onguene provided the context for the conference, including the sentiment that AI will never get smarter with static data and that using only data is not enough. (Great minds think alike!)
Cosmo Tech’s VP Global Sales Loic Gonnet was next up to unveil the program of the day. After presenting Cosmo Tech and how they use augmented intelligence instead of relying solely on data, Gonnet ensured us that we were in for an excellent afternoon of discussions and debates about artificial intelligence and the Internet of Things (IoT).The conference featured six speakers from different backgrounds, each giving an individual speech and then convening at the end for a round table discussion.
Included in the program was Luc Julia, the co-creator of Siri and the current Vice President of Innovation at Samsung Electronics. His speech was straight to the point, titled Artificial Intelligence Doesn’t Exist. He talked about how nothing is actually black box, the famous Microsoft chatbot fiasco and how autonomous vehicles will never really exist, thanks to the rule-free, close-your-eyes-and-hope-for-the-best driving techniques of the rue de l’étoile traffic circle in Paris. He ended his speech by insisting that artificial intelligence are just tools that we are providing and that we are in control, as nothing will happen until we decide.
Soon after was Cosmo Tech’s Executive Chairman and co-founder, Michel Morvan. He firmly stands behind his belief that relying on data is never enough, as what is in the future has sometimes never happened in the past. With this, he gave listeners some good news and some bad news. The bad news is that we are not able to create a machine that learns or knows how to reason, how to think. However, the good news is that there are machines that know how to reason – they’re called humans. “Just relying on data from the past not only would not be very efficient, it would not be very smart. You have a lot of people who are experts in their field, and you want to use the knowledge they have in their brain coupled with the data from the past,” he explained.
Other speakers included Dr. Daniel Covahich, Director of Big Data and AI at Braincities; Pierre Hoffer, Director HPC and AI EMEA at HPE; Alexandrea Araujo, Data Scientist at Wavestone; and Thomas Solignac, CEO of Golem.AI.
There was another theme present throughout the day: cats. As an essential building block of the internet, speakers described how artificial intelligence used cats to work on their recognition skills. While it only takes a human at most two times to recognize a cat, machines need a bit more time.
Safe to say Gonnet was correct when he told us how entertaining the afternoon would be. Attendees also received a special treat at the end – a copy of Julia’s book Artificial Intelligence Doesn’t Exist.