Rebooting AI: Building Machines we can Trust

Gary Marcus

Wednesday, September 25, 2019 at 12:00pm
545 Li Ka Shing

Artificial intelligence has a trust problem. We are relying on A.I. more and more, but it hasn’t yet earned our confidence.

Despite the intense recent hype surrounding AI, no current AI system remotely approaches the flexibility of human intelligence; as I will show, even the ability to read at grade-school level eludes current approaches.

Building on my recent synthesis with Ernest Davis, I argue that the currently popular technique of deep learning is valuable, but also, as an approach to general intelligence, radically incomplete  — and unlikely to be solved merely be gathering more data and marshaling more computational resources

I conclude with some proposals for what a new breed of artificial intelligence — informed by the cognitive sciences, and with an emphasis on common sense  — might look like.