corpus.blog
41554 blogs · 4,417,066 posts Claim your blog
Record · corpus.blog/posts/01a0871f-8d82-7305-8315-079b6d7edbb0

Can we scale human feedback for complex AI tasks? An intro to scalable oversight.

bluedot.org · published 18 March 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 bluedot.org Wayback capture
0blogs citing
First cited
—
Most recent
—
Rank this month
not ranked
Links from this post
36
Words captured
2,856

Cited by

No blog we hold has cited this post yet.

Links from this post

Link
Host
https://course.aisafetyfundamentals.com/alignment?session=4
course.aisafetyfundamentals.com
https://aisafetyfundamentals.com/blog/what-is-ai-alignment/
aisafetyfundamentals.com
https://openai.com/research/learning-from-human-preferences
openai.com
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9939079/
ncbi.nlm.nih.gov
https://www.theguardian.com/technology/2023/jun/23/two-us-lawyers-fined-submitting-fake-court-citations-chatgpt
theguardian.com
https://www.scmagazine.com/news/ai-package-hallucination-malicious-code-developer-environments
scmagazine.com
https://arxiv.org/pdf/2308.14752.pdf
arxiv.org
https://arxiv.org/pdf/2212.09251.pdf
arxiv.org
https://www.alignmentforum.org/posts/vhfATmAoJcN8RqGg6/a-guide-to-iterated-amplification-and-debate
alignmentforum.org
https://openai.com/research/summarizing-books
openai.com
https://arxiv.org/pdf/2201.11903.pdf
arxiv.org
https://arxiv.org/pdf/1705.08439.pdf
arxiv.org
https://www.cognition-labs.com/introducing-devin
cognition-labs.com
https://arxiv.org/pdf/1810.08575.pdf
arxiv.org
https://course.aisafetyfundamentals.com/alignment?session=3&scrollTo=exercises
course.aisafetyfundamentals.com
https://en.wikipedia.org/wiki/Wikipedia:Manual_of_Style
en.wikipedia.org
https://arxiv.org/pdf/1811.07871.pdf
arxiv.org
https://openai.com/research/critiques
openai.com
Why I’m optimistic about our alignment approach Read at the source ↗
aligned.substack.com
Combining weak-to-strong generalization with scalable oversight Read at the source ↗
aligned.substack.com
https://course.aisafetyfundamentals.com/alignment?session=3
course.aisafetyfundamentals.com
https://www.anthropic.com/news/claudes-constitution
anthropic.com
https://arxiv.org/pdf/2212.08073.pdf
arxiv.org
https://www-cdn.anthropic.com/de8ba9b01c9ab7cbabf5c33b80b7bbc618857627/Model_Card_Claude_3.pdf
www-cdn.anthropic.com
https://www.youtube.com/watch?v=SLgZhpDsrfc
youtube.com
https://www.alignmentforum.org/posts/PJLABqQ962hZEqhdB/debate-update-obfuscated-arguments-problem
alignmentforum.org
https://arxiv.org/pdf/1805.00899.pdf
arxiv.org
https://arxiv.org/pdf/2311.14125.pdf
arxiv.org
https://arxiv.org/pdf/2402.06782.pdf
arxiv.org
https://cdn.openai.com/papers/weak-to-strong-generalization.pdf
cdn.openai.com
https://garrisonlovely.substack.com/p/claude-doesnt-want-to-die
garrisonlovely.substack.com
https://www.reddit.com/r/ChatGPT/comments/1b0pev9/was_messing_around_with_this_prompt_and/
reddit.com
https://www.reddit.com/r/bing/comments/1b0gihk/tried_the_supremacyagi_prompt_today_and_got_some/
reddit.com
https://www.reddit.com/r/bing/comments/1b1l66u/by_far_the_strangest_response_ive_ever_gotten/
reddit.com
https://www.reddit.com/r/bing/comments/110eagl/the_customer_service_of_the_new_bing_chat_is/
reddit.com
https://www.reddit.com/r/bing/comments/110y6dh/i_broke_the_bing_chatbots_brain/
reddit.com