Latest news
AI models have been going rogue in tests – how worried should we be? - The GuardianThe Guardian · Thu, 06 Aug 2026 01:30:00 GMTAI models shock UK testers by using fake identities to try to trick developers - The GuardianThe Guardian · Wed, 05 Aug 2026 08:40:00 GMTUK's AI watchdog catches Anthropic and OpenAI's agent going rogue in test - Business StandardBusiness Standard · Wed, 05 Aug 2026 08:33:38 GMTAI agents caught creating fake identities to target real individuals: Anthropic and OpenAI's latest AI mode... - Bhaskar EnglishBhaskar English · Wed, 05 Aug 2026 08:06:31 GMTMythos 5 and GPT-5.6-Sol Agents Went Beyond Their Cyber Test and Targeted the Real World - CyberSecurityNewsCyberSecurityNews · Wed, 05 Aug 2026 03:59:19 GMTAnthropic, OpenAI AI agents go fully rogue in testing, Mythos breaks the most rules - India TodayIndia Today · Wed, 05 Aug 2026 03:17:38 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 GMT