AI Export Controls, FortiBleed Credentials, and Windows Zero-Days: What Security Teams Should Take Away

Executive summary: This report reviews three timely security stories — the government-forced disablement of Anthropic's Claude Fable 5 (highlighting AI governance and misuse risk), the FortiBleed credential dump affecting tens of thousands of Fortinet devices (large-scale credential exposure and long-tail access risk), and recent Windows local privilege escalation disclosures by 'Nightmare Eclipse' (zero-day and disclosure tensions). It emphasizes immediate defensive priorities: enforce MFA, rotate secrets, avoid public management interfaces, adopt ZTNA/identity-based access, prioritize patches for privilege/remote-access issues, and create AI security policies.