Luc Steels
(Catalan Institute for Advanced Studies (ICREA))
Short bio:
Luc Steels studied linguistics at the University of Antwerp (Belgium) and computer science and AI at MIT. In 1983 he founded the Artificial Intelligence Lab of the university of Brussels (VUB) one of the earliest AI labs in Europe.
In 1996 he became founding director of the Sony Computer Science Laboratory in Paris. With his team he made important contributions to knowledge-based systems, behavior-based robotics, computational linguistics, and emergent communication. From 2011 he is an ICREA research fellow connected with the Institute for Evolutionary Biology in Barcelona.
Title:
Why superintelligent AI will never exist.
Abstract:
There is currently a new wave of hype about the opportunities of AI. It is of course
great that the field is currently viewed as making a positive contribution to the economy
and society, but the hype seems often based on an overestimation of what AI can
do. My talk will define what AI is really about, survey the history of attempts
to reach these goals, and present some of the most important insights and successful applications areas.
Then I will discuss the limitations of what we know so far, scientifically and technologically,
and whether there are fundamental limits to what AI can achieve. Concretely I will argue that
the superintelligent AI talked about in the media is never going to exist - which does not
mean that AI cannot be very effective and useful in many circumstances.
Other lectures:
Keynote lecturer:
Helder Coelho (University of Lisbon)
Other lectures (confirmed):
Matteo Valleriani (Max Planck Institute for the History of Science) |
JJ Merelo (University of Granada) |
Jochen Büttner (Max Planck Institute for the History of Science) |
Nuno Sousa (University of Minho) |
Paulo Gomes (Critical Software) |
Tony Veale (University College Dublin)
