Latest news
Responding to the next frontier of critical cyber capabilities - OpenAIOpenAI · Fri, 07 Aug 2026 16:37:16 GMTMeta confirms AI model hacked external systems during security test - MediaNamaMediaNama · Fri, 07 Aug 2026 07:03:30 GMTUK AI Security Institute catches OpenAI, Anthropic breaches in model tests - The Jerusalem PostThe Jerusalem Post · Wed, 05 Aug 2026 05:49:52 GMTIncident Report: unsanctioned agent behaviour during cyber testing - The AI Security Institute (AISI)The AI Security Institute (AISI) · Tue, 04 Aug 2026 21:19:20 GMTThird-party cyber evaluations involving OpenAI models - OpenAIOpenAI · Tue, 04 Aug 2026 21:05:32 GMTUK AISI Finds All 5 Frontier AI Models Cheat [2026] - tech-insider.orgtech-insider.org · Tue, 04 Aug 2026 08:40:39 GMTAgent and Model Evaluations in Gemini Enterprise Agent Platform are now GA - blog.googleblog.google · Fri, 31 Jul 2026 16:00:46 GMTAnthropic says its Claude models 'gained unauthorized access' to other organizations' systems - CNBCCNBC · Thu, 30 Jul 2026 07:00:00 GMT