Surviving the AI Winter

The promise of AI miracles during the 80s was a big enticement to large corporations such as Boeing. Significant investments were made, but real breakthroughs were very rare and both time and patience ran out before some of the most interesting projects had a chance to bear fruit. However, despite the inherently conservative nature of management at all levels and the numerous gates that new ideas must pass through, an encouraging number of AI projects found their way to production usage. It is disappointing that many of those systems have since disappeared

This paper addresses the issues surrounding those systems that have succeeded in the areas of intelligent CAD, natural language understanding, and data management. It also examines some of the factors that led to the demise of some AI systems such as XCON and CASEy that looked so promising in the beginning. Both business and technical reasons are explored from an industrial perspective and some suggestions are offered to help assess the merits of future projects