corpus.blog
56,966 blogs · 4,417,066 posts Claim your blog
Record · corpus.blog/posts/01a0878e-c03c-708e-9d5a-cbbbacc73d54

Takes on “Alignment Faking in Large Language Models”

Home - Joe Carlsmith · published 18 December 2024

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
34
Words captured
18,816

Cited by

No blog we hold has cited this post yet.

Links from this post

Link
Host
https://www.anthropic.com/research/alignment-faking
anthropic.com
https://arxiv.org/abs/2311.08379
arxiv.org
https://www.lesswrong.com/posts/LJpqnbgoaTJL6iMAf/the-goal-guarding-hypothesis-section-2-3-1-1-of-scheming-ais
lesswrong.com
https://arxiv.org/abs/2308.03296
arxiv.org
https://www.lesswrong.com/posts/AFdvSBNgN2EkAsZZA/what-is-it-to-solve-the-alignment-problem-1
lesswrong.com
https://arxiv.org/abs/2411.00986
arxiv.org
On the abolition of man Read at the source ↗
joecarlsmith.com
https://console.anthropic.com/workbench/
console.anthropic.com
https://static1.squarespace.com/static/6593e7097565990e65c886fd/t/6751eb240ed3821a0161b45b/1733421863119/in_context_scheming_reasoning_paper.pdf
static1.squarespace.com
https://x.com/jkcarlsmith/status/1866232909558112508
x.com
https://www.lesswrong.com/posts/HiuagmXnhTAgDKRBP/why-focus-on-schemers-in-particular-sections-1-3-and-1-4-of
lesswrong.com
https://arxiv.org/pdf/2206.13353
arxiv.org
https://www.lesswrong.com/posts/A9YYkbnjmfsP7Chfo/a-framework-for-thinking-about-ai-power-seeking
lesswrong.com
Reality has a surprising amount of detail Read at the source ↗
johnsalvatier.org
https://deepmind.google/discover/blog/specification-gaming-the-flip-side-of-ai-ingenuity/
deepmind.google
What does Bing Chat tell us about AI risk? Read at the source ↗
cold-takes.com
Predictable updating about AI risk Read at the source ↗
joecarlsmith.com
https://arxiv.org/pdf/2311.08379
arxiv.org
https://arxiv.org/pdf/2401.05566
arxiv.org
https://arxiv.org/abs/2412.06769
arxiv.org
https://www.dwarkeshpatel.com/p/joe-carlsmith
dwarkeshpatel.com
https://arxiv.org/pdf/2112.00861
arxiv.org
https://www.anthropic.com/news/claude-character
anthropic.com
When “yang” goes wrong Read at the source ↗
joecarlsmith.com
Being nicer than Clippy Read at the source ↗
joecarlsmith.com
https://youtu.be/ugvHCXCOmm4?t=9776
youtu.be
Video and transcript of presentation on Scheming AIs Read at the source ↗
joecarlsmith.com
Otherness and control in the age of AGI Read at the source ↗
joecarlsmith.com
Does AI risk “other” the AIs? Read at the source ↗
joecarlsmith.com
https://arxiv.org/pdf/2411.00986
arxiv.org
https://github.com/redwoodresearch/alignment_faking_public/blob/master/model_organism_honly_prompts/prompt_atn_honly_with_examples.txt
github.com
https://www.transformernews.ai/p/anthropic-ai-welfare-researcher
transformernews.ai
https://nickbostrom.com/propositions.pdf
nickbostrom.com
https://www.anthropic.com/research/probes-catch-sleeper-agents
anthropic.com