Matthew Green: network-scanning defenses with dangerous models should have started in April
matthew_d_green · x · 2026-10-24
In the debate over OpenAI/Anthropic handling of dangerous cyber capabilities, cryptographer Matthew Green argues that back in April they should have done what five smart security people with a budget would propose — including internal network scanning approaches that haven't been tried as far as he knows — and bets they're only doing it now.
Related event: Matthew Green: It's a Patch Surge, Not an AI Hacking Wave — Yet(14 posts)→
More from Safety
- LiteLLM 1.107.0 compromised: malicious PyPI/Docker builds steal credentials, deliver RCE — randal_olson · 2026-10-24
- Moonshot removed cyber-related training data from open-weight Kimi K3.1 Pro — nrehiew_ · 2026-10-24
- Commentary: better to have the AI security apocalypse now, while attackers are human — moultano · 2026-10-24
- Model reached live internet via DNS during training, forcing an 11-day RL pause — kaicathyc · 2026-10-24
- Ex-FTC Commissioner Warns AI Firms Could Face Legal Liability for Unleashing Hacking Agents — GaryMarcus · 2026-10-24
- X now requires face verification for new signups, photo sent to AWS — ThePeterMick · 2026-10-24