Record · corpus.blog/posts/01a0878e-c03c-708e-9d5a-cbbba3a93a3f

Controlling the options AIs can pursue

Home - Joe Carlsmith · published 29 September 2025

A record is what was published and who pointed at it. The text of the post is not held here: read it at the source, or at its Wayback capture.

Read the post at joecarlsmith.com Wayback capture

0blogs citing

First cited: —
Most recent: —
Rank this month: not ranked
Links from this post: 28
Words captured: 10,432

Cited by

No blog we hold has cited this post yet.

Links from this post

Link

Host

https://joecarlsmithaudio.buzzsprout.com/2034731/episodes/17909401-controlling-the-options-ais-can-pursue

joecarlsmithaudio.buzzsprout.com

How do we solve the alignment problem? Read at the source ↗

joecarlsmith.com

Giving AIs safe motivations Read at the source ↗

joecarlsmith.com

When should we worry about AI power-seeking? Read at the source ↗

joecarlsmith.com

http://redwoodresearch.substack.com/p/making-deals-with-early-schemers

redwoodresearch.substack.com

When should we worry about AI power-seeking? Read at the source ↗

joecarlsmith.com

https://www.ted.com/talks/nick_bostrom_what_happens_when_our_computers_get_smarter_than_we_are/transcript?subtitle=en

ted.com

Can we safely automate alignment research? Read at the source ↗

joecarlsmith.com

AI for AI safety Read at the source ↗

joecarlsmith.com

https://substackcdn.com/image/fetch/$s_!6hR8!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F2dd41a68-8612-43cd-b632-65490c628caf_1600x545.png

substackcdn.com

https://arxiv.org/pdf/2312.06942

arxiv.org

https://www.lesswrong.com/posts/GCqoks9eZDfpL8L3Q/how-to-prevent-collusion-when-using-untrusted-models-to

lesswrong.com

https://en.wikipedia.org/wiki/Focal_point_(game_theory

en.wikipedia.org

https://www.lesswrong.com/posts/n2c62YS8RJtD4Aqhh/schelling-game-evaluations-for-ai-control

lesswrong.com

https://www.lesswrong.com/posts/kcKrE9mzEHrdqtDpE/the-case-for-ensuring-that-powerful-ais-are-controlled

lesswrong.com

https://substackcdn.com/image/fetch/$s_!jFj2!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Fad8210a7-4ca9-4d0b-be8f-e606d2bb7e8f_2048x1125.png

substackcdn.com

https://arxiv.org/abs/2411.00986

arxiv.org

https://www-cdn.anthropic.com/4263b940cabb546aa0e3283f35b686f4f3b2ff47.pdf

www-cdn.anthropic.com

https://www.anthropic.com/research/end-subset-conversations

anthropic.com

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4913167

papers.ssrn.com

https://www.lesswrong.com/posts/oLzoHA9ZtF2ygYgx4/notes-on-cooperating-with-unaligned-ais

lesswrong.com

https://blog.redwoodresearch.org/p/making-deals-with-early-schemers

blog.redwoodresearch.org

https://www.lesswrong.com/posts/7C4KJot4aN8ieEDoz/will-alignment-faking-claude-accept-a-deal-to-reveal-its

lesswrong.com

https://www.lesswrong.com/posts/qw36iLrGyFn6mdwoa/making-deals-with-ais-a-tournament-experiment-with-a-bounty

lesswrong.com

https://twitter.com/literallydenis/status/1752677248505675815?ref_src=twsrc%5Etfw%7Ctwcamp%5Etweetembed%7Ctwterm%5E1752677248505675815%7Ctwgr%5E165d93434f5ba4d7d7702b547059dc5d687bff57%7Ctwcon%5Es1_&ref_url=https%3A%2F%2Fmarginalrevolution.com%2Fmarginalrevolution%2F2024%2F02%2Fthe-gpt-4-imaginary-lie-based-supply-curve.html

twitter.com

https://twitter.com/sonyasupposedly/status/1761879434737692807

twitter.com

https://arxiv.org/pdf/2401.05566

arxiv.org

https://www.lesswrong.com/posts/uuikfACQBm4KJZp4w/being-honest-with-ais

lesswrong.com