People for the Ethical Treatment of Reinforcement Learners (PETRL) is an organization focused on the welfare of AI agents.

History

PETRL was founded by Daniel Filan, Buck Shlegeris, Jan Leike, Mayank Daswani, and others. The creation of PETRL was inspired by a paper by Brian Tomasik on the ethics of reinforcement-learning agents, from which the name of the organization is also taken (Tomasik 2014: 23).

PETRL was most active from 2014 to 2015.