factual
Interpretability agent wins auditing game end-to-end
An Interpretability Agent—a version of Claude equipped with interpretability tools like a 'get top active features' call—can autonomously win the auditing game end-to-end: it converses with a deliberately misaligned model, finds the subtle underlying evil behavior, and systematically validates and explores its downstream effects, matching what human teams did (one of which won in 90 minutes against a three-day budget).
factualpending
Speaker
Trenton BrickenEvidence Quote
“It is also able to win the auditing game and discover the bad behavior. End to end, yeah. You give it the same prompt that the humans had. You fire it off.”
Created: 6/13/2026, 3:28:51 AM
Your notes
Loading notes...