It is the intent of this article to push beyond the horizon of "current survey and future work," with a set of mildly outrageous long-term challenges. The targeted areas are expert systems, problem solving and learning. Somewhere between AAAI-82 and science fiction, our future research projects lie. If we are not to be mired in the pedestrian, we should reach out beyond the traditional. Not long ago, a proposal to develop heuristics (Lenat, 1982) or follow strategic instructions (Mostow) would have seemed mildly outrageous. This article is intended as a set of equally unreasonable challenges, a focus for the future. Expert systems, problem solving and learning are particularly amenable to such a focus. The categories are neither mutually exclusive nor collectively exhaustive.
/lp/association-for-computing-machinery/challenges-5VByJ9dSRq