Discuss Read More
Related Posts
[Paper] Output Supervision Can Obfuscate the CoT
Published on November 20, 2025 10:41 PM GMTWe show that training against a monitor that…
Response to Introspective Awareness research
Published on December 19, 2025 5:23 PM GMTThis is a rewrite of a comment I…
Blocking live failures with synchronous monitors
A common element in many AI control schemes is monitoring – using some model to review…