Full Width [alt+shift+f] Shortcuts [alt+shift+k]
Sign Up [alt+shift+s] Log In [alt+shift+l]
16
A new Anthropic paper reports that reasoning model chain of thought (CoT) is often unfaithful. They test on Claude Sonnet 3.7 and r1, I’d love to see someone try this on o3 as well.
a month ago

Improve your reading experience

Logged in users get linked directly to articles resulting in a better reading experience. Please login for free, it takes less than 1 minute.

More from Don't Worry About the Vase

AI #113: The o3 Era Begins

Enjoy it while it lasts.

a week ago 14 votes
o3 Is a Lying Liar

I love o3.

2 weeks ago 15 votes
You Better Mechanize

Or you had better not.

2 weeks ago 13 votes
Crime and Punishment #1

This seemed like a good next topic to spin off from monthlies and make into its own occasional series.

2 weeks ago 12 votes
o3 Will Use Its Tools For You

OpenAI has finally introduced us to the full o3 along with o4-mini.

2 weeks ago 12 votes

More in AI

AI #113: The o3 Era Begins

Enjoy it while it lasts.

a week ago 14 votes
OpenAI’s dirty December o3 demo doesn’t readily replicate

Don’t believe everything you see

2 weeks ago 16 votes
o3 Is a Lying Liar

I love o3.

2 weeks ago 15 votes
New Results of State-of-the-art LLMs on 4 Political Orientation Tests

One model appears closer to the center than the rest

2 weeks ago 14 votes
You Better Mechanize

Or you had better not.

2 weeks ago 13 votes