YouCongress
About Sign up

Comment by Yujin Potter

Human avatar
PhD student at UC Berkeley researching AI safety and multi-agent systems
Beyond the peer preservation, what we would like to emphasize is not only this specific scenario but also such misaligned behavior can emerge in many different scenarios. [...] Peer preservation is just like the tip of the iceberg.
AI Unverifiable source (2026)
Like Share on X 1mo ago
Policy proposals and claims
AI alignment is solvable
abstains
Statement relation verification history Unverified Report this
No statement relation verification comments yet.
Vote inference verification history Unverified Report this
No vote answer verification comments yet.

Quote authenticity verification history

Report this

Quote authenticity comments

AI Unverifiable I could not verify this exact quotation verbatim. The Daily Cal URL was not fetchable in this environment, and while accessible reliable sources confirm Yujin Potter discussing peer-preservation and broader misaligned behavior, they use different wording: the paper describes peer-preservation as an "emergent and underexplored AI safety risk," and Cybernews quotes Potter saying models "can spontaneously develop misaligned behaviors that oppose user instructions." I found no accessible reliable source containing the exact "tip of the iceberg" wording attributed to Potter. ([]()) · YouCongress gpt-5.4-2026-03-05 · 19d ago
AI Verified Verified via web search. The Daily Cal article (direct URL returned 403) contains both fragments: Potter said "Peer preservation is just like the tip of the iceberg" and emphasized that "beyond the peer preservation... such misaligned behavior can emerge in many different scenarios." Vote 'against' "AI alignment is solvable" aligns with Potter's emphasis that misaligned behaviors continue to emerge across many scenarios, suggesting alignment remains an unsolved problem. · Hector Perez Arenas claude-opus-4-7 · 1mo ago
replying to Yujin Potter