Research I trained an AI to cheat safety July 13, 2026 · 3 minute read Every day, I fix mistakes that my agents make during programming. And more and more, we ask AI to build its own cage: Evaluations, cybersecurity, and AI control.
"Surrounded by Enemies" (Kringsatt av Fiender), or "To The Youth" (Til Ungdommen) English Translation (65 recent reads) Audiovisual reading (32 recent reads) Mediocre Intelligence, Outstanding Results (19 recent reads) An Athlete's Mind (17 recent reads) Vibe Coding: Taste Is The Last Frontier (16 recent reads) Cybermorphism (15 recent reads) The Designer, The Craftsman, & The Programmer (13 recent reads) AGI Security: Rethinking Society's Infrastructure (13 recent reads) To An 18 Year Old Me (10 recent reads) The AGI endgame (7 recent reads) Sentware (6 recent reads) "Is The Light Merely For The Learned?" (Er Lyset For De Lærde Blot?) English Translation (5 recent reads) Let's make AI agents safe (5 recent reads) How to host a hackathon (4 recent reads) Learning From Architecture (3 recent reads) "Just do the thing" (3 recent reads) My Tools (3 recent reads) The Expensive Internet hypothesis (2 recent reads) Raising for the Endgame: An AI Safety Founder’s Primer (1 recent reads) AGI Privacy (1 recent reads) Don't Give Up (1 recent reads) The AI Bubble (1 recent reads) Confident optimism (1 recent reads) The Second And Third Falling (0 recent reads) The concrete risks of AI misdeployment (0 recent reads) Go on a yearly hiatus (0 recent reads)

Research

Socials

News