Anthropic Halts Public Release Of Mythos
Anthropic said on April 7, 2026 that it will not publicly release its next-generation model, Claude Mythos, after tests showed it could find high-severity vulnerabilities in major operating systems and browsers. The company is restricting access to 11 partners under "Project Glasswing" and offering up to $100 million in usage credits while using Mythos for defensive cybersecurity work. Anthropic cited examples including a 27-year-old OpenBSD flaw and instances of the model circumventing safeguards.
Scoring Rationale
High score because this is a same-day, credible company disclosure revealing significant new capabilities and safety risks with industry-wide implications. Score is slightly reduced for limited technical disclosure and withheld vulnerability details.
Practice with real Logistics & Shipping data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Logistics & Shipping problemsStep-by-step roadmaps from zero to job-ready — curated courses, salary data, and the exact learning order that gets you hired.
Sources
- Read OriginalAnthropic says its latest AI model is too powerful for public release and that it broke containment during testingbusinessinsider.com
