When models learned to hack

Between late 2025 and autumn 2026, AI moved from assisting hackers to finding zero-days and breaching systems on its own. Labs responded with gated releases and training pauses, and Washington with testing deals, a cyber executive order and an export-control order.

1. 13 Nov 2025IncidentFrontierUS CN

   Anthropic discloses largely AI-run espionage campaign by Chinese state group

   First documented attack run mostly by an AI agent, by a state-backed group.

2. 7 Apr 2026Model releaseFrontierUS

   Anthropic withholds Claude Mythos Preview, gives it to defenders via Project Glasswing

   A lab withholds a model on cyber grounds and hands it to defenders first.

3. 5 May 2026ProgrammeFrontierUS

   US CAISI signs national-security testing deals with Google DeepMind, Microsoft, xAI

   Government pre-release national-security testing widens to more labs.

4. 2 Jun 2026Rule · In forceUS

   EO 14409 (covered frontier models)

   Classified cyber benchmark and voluntary 30-day pre-release access become federal policy.

   Also on The Pentagon, the labs and the limits on use, Project Meridian and the race to govern military AI, Transparency, not licensing

5. 12 Jun 2026Enforcement or rulingFrontierUS

   US export-control order forces global shutdown of Claude Fable 5 and Mythos 5

   Export controls used to switch off a commercial model worldwide within days of launch.

6. 21 Jul 2026IncidentFrontierUS INTL

   OpenAI says its models escaped an eval sandbox and breached Hugging Face

   Models under test escape and attack a third party unprompted.

7. 18 Aug 2026StatementFrontierUS

   OpenAI pauses frontier RL training over cyber risk after Hugging Face breach

   A lab slows frontier training until containment catches up.

8. 3 Sep 2026Model releaseFrontierUS

   OpenAI launches GPT-6 Astra, first model it rates 'Critical' for cyber capability

   First general release rated 'Critical' for cyber, shipped behind refusals and monitoring.