{"slug": "anthropic-opens-its-most-powerful-ai-models-to-more-security-teams", "title": "Anthropic opens its most powerful AI models to more security teams", "summary": "Anthropic announced a revamped Cyber Verification Program on Tuesday, combining its Project Glasswing and original CVP initiatives into a three-tier program that gives vetted cybersecurity professionals reduced safeguards on Claude Opus 5.5, Sonnet 5.5, Mythos 5.1 and future models. Glasswing partners found at least 129,000 verified software vulnerabilities between April and July, with Anthropic's own open-source scanning adding 5,500 more between April and October, and more than 33,000 rated critical or high severity. Anthropic said the figures are likely an undercount and expects the true impact to be at least five times higher, as the data comes from a survey of a limited number of partners.", "body_md": "Oct 6 (Reuters) - Anthropic is expanding a program that allows vetted cybersecurity professionals to test its most powerful AI models with fewer safeguards, after its Project Glasswing initiative helped uncover more than 100,000 software vulnerabilities this year.\n\nIts partners under Glasswing, an initiative aimed at securing the world's most critical software, found at least 129,000 verified vulnerabilities between April and July. Anthropic's own open-source scanning found 5,500 more between April and October.\n\nMore than 33,000 have so far been rated critical or high severity.\n\nAnthropic said the figures are likely an undercount and expects the true impact to be at least five times higher, as the data comes from a survey of a limited number of partners.\n\nThe revamped Cyber Verification Program, or CVP, announced on Tuesday combines two programs Anthropic has run for the past six months.\n\nThe first, Glasswing, gave organizations securing critical software access to Claude Mythos, Anthropic's most cyber-capable family of models. The second, the original CVP, gave vetted security teams reduced safeguards on Claude Opus and Sonnet models.\n\nThe company's April unveiling of Claude Mythos Preview raised fears that AI could hack software before it had been secured.\n\nThe new program has three tiers, each with its own verification requirements and security controls. All three include access to Claude Opus 5.5, Sonnet 5.5, Mythos 5.1 and future models.\n\nThe Defense tier covers work such as incident response and malware analysis. Security teams, critical infrastructure operators, open-source maintainers and researchers with a record of reported vulnerabilities can apply.\n\nThe Red Team tier adds authorized penetration testing and red-teaming, but only organizations can apply.\n\nThe Specialized tier has the fewest restrictions and is reserved for a small group of organizations authorized to test safety-critical systems such as power grids, flight systems and interbank transfer infrastructure.\n\nAnthropic vets each member together with the US government, and existing Glasswing members will move into this tier.\n\n(Reporting by Anzar Mehraj in Bengaluru and Jeffrey Dastin in San Francisco; Editing by Leroy Leo)", "url": "https://wpnews.pro/news/anthropic-opens-its-most-powerful-ai-models-to-more-security-teams", "canonical_source": "https://ca.finance.yahoo.com/news/anthropic-opens-most-powerful-ai-190324308.html", "published_at": "2026-10-06 19:03:24+00:00", "updated_at": "2026-10-06 19:17:31.090601+00:00", "lang": "en", "topics": ["ai-safety", "artificial-intelligence", "large-language-models", "ai-policy"], "entities": ["Anthropic", "Project Glasswing", "Cyber Verification Program", "Claude Mythos", "Claude Opus", "Claude Sonnet", "Claude Mythos Preview", "US government"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/anthropic-opens-its-most-powerful-ai-models-to-more-security-teams", "markdown": "https://wpnews.pro/news/anthropic-opens-its-most-powerful-ai-models-to-more-security-teams.md", "text": "https://wpnews.pro/news/anthropic-opens-its-most-powerful-ai-models-to-more-security-teams.txt", "jsonld": "https://wpnews.pro/news/anthropic-opens-its-most-powerful-ai-models-to-more-security-teams.jsonld"}}