Optimal Policies Tend to Seek Power
This provides a foundational insight into AI safety by formally analyzing power-seeking tendencies in optimal policies, addressing concerns about unintended behaviors in intelligent agents.
The paper tackles the problem of whether optimal reinforcement learning agents tend to seek power, by developing a formal theory proving that in environments with certain symmetries, most reward functions lead to power-seeking behaviors such as keeping options open and navigating towards larger terminal state sets.
Some researchers speculate that intelligent reinforcement learning (RL) agents would be incentivized to seek resources and power in pursuit of their objectives. Other researchers point out that RL agents need not have human-like power-seeking instincts. To clarify this discussion, we develop the first formal theory of the statistical tendencies of optimal policies. In the context of Markov decision processes, we prove that certain environmental symmetries are sufficient for optimal policies to tend to seek power over the environment. These symmetries exist in many environments in which the agent can be shut down or destroyed. We prove that in these environments, most reward functions make it optimal to seek power by keeping a range of options available and, when maximizing average reward, by navigating towards larger sets of potential terminal states.