2015 beneficial-AI open letter

In January 2015, the Future of Life Institute brought together AI researchers and experts in economics, law and ethics at its Puerto Rico conference. Its year-end account identifies the beneficial-AI letter and companion agenda as products of that meeting. This entry uses month precision for the initiative; the current letter webpage displays October 28, 2015, a later webpage date that does not establish first release.^[2][1]

What the letter endorsed

The letter advocates research into making increasingly capable AI serve human purposes. Its central formulation is “our AI systems must do what we want them to do” (recommendation paragraph). The proposed work spans technical disciplines and economics, law and philosophy.^[1]

The original lists Stuart Russell, Geoffrey Hinton, Yoshua Bengio and Yann LeCun among prominent signatories. Signing supports this letter’s research recommendation; it does not establish shared forecasts, risk probabilities or endorsement of later letters.^[1]

Academic context and limits

The companion research agenda explains the technical questions behind the appeal. The letter documents an advocacy initiative, not an AI failure or an experimental safety result. Its statement of agreement is the signatories’ position, rather than a representative survey of the field.

Related public discussion

Science Friday’s April 10, 2015 segment lists Stuart Russell, Eric Horvitz and Max Tegmark as guests discussing AI’s future. This verifies a related public discussion, without establishing what each guest said about the letter: the retrieved page contains no transcript body and its recording has not been checked.^[4]