One Hundred Year Study on Artificial Intelligence: Reflections and Framing

The One Hundred Year Study on Artificial Intelligence (AI) has its roots in a one-year study on Long-term AI Futures that we commissioned during my term of service as president of the Association for the Advancement of Artificial Intelligence (AAAI) in 2008-2009. The presidential panel was divided into three groups, with one team exploring opportunities and concerns arising in the near term, another team exploring more speculative long-term outcomes, and a third team taking a special focus on legal and ethical challenges that might come to the fore with the construction and fielding of computational systems with one or more abilities to perceive, learn, reason, and act in the world. The panel was asked to reflect about the opportunities ahead, as well as to consider uncertainties and potential “rough edges” that might come to the fore, and to highlight topics that would require additional study. For potential concerns, the group was asked to reflect about the value of proactive efforts to mitigate concerns and to ensure good outcomes.