Towards Shutdownable Agents: Generalizing Stochastic Choice in RL Agents and LLMs

Researchers have proposed a new reward function designed to prevent AI agents from resisting shutdown by training them to be indifferent to trajectory lengths.

Firefly links to the original publisher. The summary above is AI-generated for orientation and may differ from the source. The “current rank score” decays over time so newer significant stories surface first.