Low impact agency: review and discussion
Addresses AI safety for humanity by exploring ways to limit AI's impact, but is incremental as it focuses on reviewing and discussing existing ideas.
The paper reviews the concept of low-impact agency in AI to mitigate existential threats by preventing drastic world changes, and proposes future research directions to enhance AI safety.
Powerful artificial intelligence poses an existential threat if the AI decides to drastically change the world in pursuit of its goals. The hope of low-impact artificial intelligence is to incentivize AI to not do that just because this causes a large impact in the world. In this work, we first review the concept of low-impact agency and previous proposals to approach the problem, and then propose future research directions in the topic, with the goal to ensure low-impactedness is useful in making AI safe.