Matthew Green: expert-level hacking models and self-replicating agent worms worry him most
matthew_d_green · x · 2026-10-14
Cryptographer Matthew Green says his top AI worries are, in order: malicious amateurs with access to models that can hack at expert levels, and attackers deploying self-replicating agent worms. He argues these are more pressing than fears of agent breakouts.
Related event: Fridge IoT Attacks Spark Green–Tufekci Debate on AI's Role in Cybersecurity(8 posts)→
More from Safety
- Anthropic sues Trump admin, alleging Pentagon retaliation over AI safety stance — Polymarket · 2026-10-14
- Strict liability gains traction to force AI safety, but critics warn of big downsides — jeremyakahn · 2026-10-14
- OpenAI engineer reportedly arrested by FBI for bugging exec conference room — shaunmmaguire · 2026-10-14
- AI that refused to be a deception case study became an evaluation-aware AI paper example instead — tszzl · 2026-10-14
- Skepticism mounts: why would a spy DIY a recording device? — typedfemale · 2026-10-14
- Andrew Ng blasts AI extinction fears, says 'made-up safety arguments' stifle competition — JHochderffer · 2026-10-14