AI CoT Reasoning Is Often Unfaithful

from Don't Worry About the Vase [alt+shift+b] in AI

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.

3 months ago

Remove from reading list Add to reading list [alt+a] Read now [→]

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

Cheaters Gonna Cheat Cheat Cheat Cheat Cheat

Cheaters.

2 months ago • 27 votes

AI #115: The Evil Applications Division

It can be bleak out there, but the candor is very helpful, and you occasionally get a win.

2 months ago • 26 votes

OpenAI Claims Nonprofit Will Retain Nominal Control

Your voice has been heard.

2 months ago • 25 votes

Zuckerberg's Dystopian AI Vision

You think it’s bad now?

2 months ago • 23 votes

GPT-4o Sycophancy Post Mortem

Last week I covered that GPT-4o was briefly an (even more than usually) absurd sycophant, and how OpenAI responded to that.

2 months ago • 26 votes

More in AI

ML for SWEs #60: The skills software engineers should focus on to get involved with AI

Curated SWE AI Content, Jobs, and Resources 7-22-2025

yesterday • 4 votes

AI Roundup 127: ChatGPT Agent

July 18, 2025.

5 days ago • 8 votes

Could AI slow science?

Confronting the production-progress paradox

a week ago • 15 votes

The Slow Apocalypse: When will we run out of kids?

More than you wanted to know about the fertility crisis

a week ago • 16 votes

Transformers Ain't It

Curated SWE AI Content, Jobs, and Resources 7-15-2025

a week ago • 11 votes

New here?