Blog
Notes on AI security and safety research. RSS feed
2026
Future of AI Honeypots
As AI agents get better, I expect them to recognize honeypots and avoid fingerprinting, so honeypots will have to rely on the combination of intelligence and speed.
12 November · 5 min read
10 prompts to protect yourself in the age of AI
10 copy-paste prompts for your AI agent to harden yourself.
10 November · 5 min read
Security has always been about future threats
Designing defenses around the future capabilities of attackers has been normal security practice for decades.
14 October · 6 min read
Why I expect AI replication incidents by 2027
I think a major incident of autonomous AI replication in the wild before the end of 2027 is reasonably likely.
27 September · 5 min read · LessWrong
2025
Hunting for AI Hackers: LLM Agent Honeypot
Our current results indicate that LLM hacking agents exist but are in the very early stages of technology adoption for mass hacking.
12 February · 7 min read · with Jacob Haimes · Apart Research
Predicting AI Releases Through Side Channels
I recently explored whether we could predict major AI releases by analyzing the Twitter activity of OpenAI's red team members.
7 January · 2 min read · LessWrong