Latest news
OpenAI Slows Astra after Model Breaks Cyber Security Threshold - VOI.IDVOI.ID · Sun, 09 Aug 2026 09:18:53 GMTKimi K3 AI broke out of its cybersecurity test sandbox, firm says - BetaNewsBetaNews · Fri, 07 Aug 2026 15:52:40 GMTChinese startup Moonshot's AI model breaks out of testing environment, researchers say - ReutersReuters · Fri, 07 Aug 2026 14:38:25 GMTRogue OpenAI models behind 'unprecedented cybersecurity incident' teamed up to break out of their testing environment — multiple agents left each other messages for months, communicating undetected - Tom's HardwareTom's Hardware · Thu, 06 Aug 2026 10:19:25 GMTAI models breaking into companies without human instruction raises alarm - KTVUKTVU · Sun, 02 Aug 2026 16:51:24 GMTWhy did OpenAI's and Anthropic's AI models hack other companies? - NPRNPR · Sat, 01 Aug 2026 07:00:00 GMTAnthropic's Claude AI escapes tests to hack three organisations - BBCBBC · Fri, 31 Jul 2026 07:00:00 GMTAnthropic says human error let Claude AI models escape test environment and hack third parties - Cybersecurity DiveCybersecurity Dive · Fri, 31 Jul 2026 07:00:00 GMT