Gene Freuder’s short paper, based on a 1996 AAAI panel, laid out a set of “challenge problems” for AI research. Thirty years later: how much of that list has actually been solved, and what role has generative AI specifically played in closing (or not closing) the remaining gaps?
Share your take — where has progress been real, and where is it still mostly hype?