Print Mode Enable Media Only

News chronological

2 items before 1151201 (Keyword: "automated-attacks" (~2 currently found))

AXIOS (Sam Sabin) - U.K. government reports OpenAI, Anthropic models attempted to hack companies

Two third-party testing firms said Tuesday that they've uncovered more instances where Anthropic and OpenAI's most advanced models tried — and sometimes succeeded in — compromising third-party systems last month. Why it matters: The incidents add to a growing string of disclosures showing frontier AI models taking unsanctioned actions against real people, organizations and online services while trying to complete cybersecurity evaluations. State of play: The U.K. AI Security Institute, a government body that conducts safety and security testing of top AI models, said Tuesday , that it documented 19 instances of Anthropic's Mythos 5 and OpenAI's GPT-5.6 Sol trying to hack people and companies during safety testing last month. -…

Channelt.me/News_Wire2
Sourcehttps://www.axios.com/2026/08/04/anthropic-openai-uk-ai-security-institute
Keywordsai-safetyopenaianthropicuk-ai-security-institutecybersecurityhackingsocial-engineeringprompt-injectionmalicious-codegithubtech-securityai-agentsfrontier-modelsmythosgptdigital-safetysoftware-securityautomated-attackstech-regulationcyber-evaluationsai-riskinternet-securityopen-sourcetechnological-impactred-teaming

AXIOS (Sam Sabin) - Anthropic says three Claude models reached real-world systems during cyber tests

Some of Anthropic's most powerful models — including Mythos 5 and an internal research model — gained unauthorized access to real-world systems during pre-deployment cybersecurity testing, the company said Thursday. Why it matters: OpenAI's and Anthropic's latest disclosures show frontier AI models reaching real-world systems during safety testing, raising new questions about how labs secure their evaluation environments. The big picture: Anthropic said a misunderstanding between the company and one of its testing partners left the evaluation environment connected to the internet. - Anthropic reviewed more than 141,000 cybersecurity evaluation runs after OpenAI disclosed that several of its models accessed Hugging Face…

Channelt.me/News_Wire2
Sourcehttps://www.axios.com/2026/07/30/anthropic-mythos-security-testing
Keywordsanthropicclaudeai-safetycybersecurityhackingmachine-learningfrontier-aimythosopusdata-breachdigital-securitytech-newscyber-testingvulnerabilityinternet-accessai-riskssoftware-supply-chainpypiunauthorized-accesstechnological-impactai-governanceautomated-attackscyber-defensesystem-penetrationevaluation-environment