{"id":460321,"date":"2026-05-26T06:21:11","date_gmt":"2026-05-26T06:21:11","guid":{"rendered":"https:\/\/www.newsbeep.com\/il\/460321\/"},"modified":"2026-05-26T06:21:11","modified_gmt":"2026-05-26T06:21:11","slug":"anthropic-says-mythos-has-uncovered-over-10k-vulnerabilities","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/il\/460321\/","title":{"rendered":"Anthropic Says Mythos Has Uncovered Over 10K Vulnerabilities"},"content":{"rendered":"

Anthropic<\/a> says its Mythos Preview model has ferreted out more than 10,000 cybersecurity vulnerabilities.<\/p>\n

That\u2019s according to a report<\/a> issued last week by the artificial intelligence startup on Project Glasswing, its effort to use AI to counter AI-powered cyberattacks.<\/p>\n

\u201cProgress on software security used to be limited by how quickly we could find new vulnerabilities,\u201d the report said. \u201cNow it\u2019s limited by how quickly we can verify, disclose, and patch the large numbers of vulnerabilities found by AI.\u201d<\/p>\n

Launched in April<\/a>, Project Glasswing is a joint venture by Anthropic and 50 partner organizations. Since the project began, their work has found more than 10,000 \u201chigh- or critical-severity vulnerabilities\u201d in \u201cthe most systemically important software in the world,\u201d the report added.<\/p>\n

In addition, several of these companies have reported that their rate of bug-finding has risen by more than a factor of 10, Anthropic said.\u00a0<\/p>\n

The startup cites the example of Cloudflare<\/a>, which has found 2,000 bugs (400 of them high- or critical-severity) within its critical-path systems,\u201cwith a false positive rate that Cloudflare\u2019s team considers better than human testers.\u201d<\/p>\n

Advertisement: Scroll to Continue<\/p>\n

The report also noted that the rate of AI progress means that many other AI companies will soon develop their own models<\/a> on the same level as Mythos Preview. However, no company has created protections that can keep these models from being misused, Anthropic added, which is why it has yet to make Mythos-class models publicly available.<\/p>\n

\u201cBut it\u2019s also why we began Project Glasswing: If a similarly capable model is released without such safeguards, it will soon become dramatically cheaper and easier for almost anyone in the world to exploit flawed software,\u201d Anthropic said.<\/p>\n

Research by PYMNTS Intelligence and Trulioo<\/a> shows that large enterprises are increasingly dealing with AI-powered cyberattacks.<\/p>\n

\u201cLarger firms, with their larger footprints, can be more susceptible<\/a> to the AI-powered spoofing of identity documents thanks to the industrialization of deepfakes and automated data scraping capabilities by adversarial cyber actors,\u201d PYMNTS wrote last week.<\/p>\n

The research found that most companies<\/a> (58%) with more than $1 billion in annual revenue surveyed reported dealing with AI-generated documents or deepfake-related attacks in the past year, a full 11 percentage points more than smaller firms. Automated scraping attacks also increased sharply with scale.<\/p>\n

\u201cAgainst this backdrop, enterprise-grade firms increasingly depend on identity systems not only to stop fraudsters, but also to determine whether legitimate customers can transact, onboard, access services and remain engaged,\u201d PYMNTS wrote. \u201cIn that environment, every authentication decision becomes a revenue decision.\u201d<\/p>\n","protected":false},"excerpt":{"rendered":"Anthropic says its Mythos Preview model has ferreted out more than 10,000 cybersecurity vulnerabilities. That\u2019s according to a…\n","protected":false},"author":2,"featured_media":460322,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[345,214463,214464,1463,343,344,4287,85,46,187484,43,214465,262,125,264],"class_list":["post-460321","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-ai-cyberattacks","tag-ai-cybersecurity","tag-anthropic","tag-artificial-intelligence","tag-artificialintelligence","tag-cybersecurity","tag-il","tag-israel","tag-mythos","tag-news","tag-project-glasswing","tag-pymnts-news","tag-technology","tag-whats-hot"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/posts\/460321","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/comments?post=460321"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/posts\/460321\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/media\/460322"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/media?parent=460321"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/categories?post=460321"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/tags?post=460321"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}