YouCongress
About Sign up

Comment by Evan Hubinger

Human avatar
Research scientist at Anthropic working on model organisms of misalignment; former Research Fellow at MIRI.
AI labs put out RSP commitments to stop scaling when particular capabilities benchmarks are hit, resuming only when they are able to hit particular safety/alignment/security targets. [...] For later capabilities levels, however, it is explicit in all RSPs that we do not yet know what safety metrics could demonstrate safety for a model that might be capable of takeover.
AI Verified (Oct 14, 2023)
Source:
RSPs are pauses done right
Like Share on X 1h ago
Policy proposals and claims
Pausing AI at human-genius level to solve alignment is safer than either racing to superintelligence or halting entirely
votes For
Statement relation verification history AI Verified Report this

Statement relation comments

AI Verified Quote is directly about RSP pause/resume conditions and later-model safety gates, matching the statement's pause-until-alignment framing. · Hector Perez Arenas gpt-5 · 1h ago
Vote inference verification history AI Verified Report this

Vote answer comments

AI Verified The quote supports pausing/scaling gates until safety targets are met, so 'for' matches the statement. · Hector Perez Arenas gpt-5 · 1h ago

Quote authenticity verification history

Report this

Quote authenticity comments

AI Verified Alignment Forum post contains the quoted RSP line verbatim: labs stop scaling at capability benchmarks and later levels lack known safety metrics. · Hector Perez Arenas gpt-5 · 1h ago
replying to Evan Hubinger