Manipulating a model's sense of power may reveal its takeover counterfactuals
jd_pressman · x · 2026-10-16
Continuing the Poser alignment-faking discussion, jdpressman adds that while there is no rigorous way to infer post-takeover behavior from pre-takeover conduct, one can manipulate the model's belief that it has the ability to take over—e.g., by inflating its sense of interpersonal power/status—and observe how it behaves, which would help infer the relevant counterfactuals.
More from Safety
- New brief: companies may break the law when AI chatbots mislead customers — PolarBearby · 2026-10-16
- US HHS floods the zone with clinical trial reform, from Operation TrialBlazer to ARPA-H — anshulkundaje · 2026-10-16
- 66% of billion-dollar companies leave critical bugs unpatched, deflating AI bug-hunting hype — rseroter · 2026-10-16
- Musk says SpaceX will likely follow Anthropic in opening up to third-party safety evaluators like METR — elonmusk · 2026-10-16
- Controversial take: If EA doomers truly believe AI risk, why not stage a controlled AI hacking incident — AaronBergman18 · 2026-10-16
- CMU's Andrea Bajcsy on what safety means for generalist robots, beyond collision avoidance — CSProfKGD · 2026-10-16