Research
Research.
This page is writing from work I’m in, not a blog.
August 27, 2026
The Adversarial CouncilA pre-registered benchmark: do two AI models catch more bugs than one? A negative result, run honestly.
Research
This page is writing from work I’m in, not a blog.
August 27, 2026
The Adversarial CouncilA pre-registered benchmark: do two AI models catch more bugs than one? A negative result, run honestly.