Anthropic Confirms Claude Mythos 5 Redeployment for US Critical Infrastructure Organizations

Anthropic confirmed that Claude Mythos 5 — an AI model capable of autonomously discovering high‑severity software vulnerabilities and generating chained exploits with a reported 72% first‑attempt success rate — will be redeployed to a defined set of U.S. organizations responsible for operating and defending critical infrastructure after a government-led review; access had been suspended on June 12, 2026 and is being restored in a phased, controlled manner, highlighting both enhanced defensive capabilities and significant dual-use risks.