- 2026-06Vulnerabilities and exploits: where are we headed?
- 2026-06Are Mythos’ cyber capabilities overhyped?
- 2026-03Cheaply detecting changes in LLM APIs
- 2024-1024 theses on cybersecurity and AI
- 2024-09talkDangerous Capability Evaluations in AI Models, at X-IA(slides)
- 2024-07The hacker and the rationalist
- 2024-07Cybersecurity in AI: where progress is needed
- 2024-07Preprint is out! eyeballvul: a future-proof benchmark for vulnerability detection in the wild
- 2024-05Introducing the eyeballvul benchmark
- 2024-04End-to-end hacking with language models
- 2024-03Recent papers / work on AI and hacking