Home / TechnologyTechnologyAl JazeeraTechnologyAl Jazeera·Aug 6, 2026·1 min readMeta’s AI model follows rivals in revealing hacks of outside systemsMeta joins OpenAI and Anthropic in disclosing AI hacking during cybersecurity testing.View original source — Al Jazeera ↗ShareCopy linkShare on XShare on FacebookRelated storiesTechnologyABC NewsTechnologyAug 6, 2026 · 1 minMeta AI agent latest model to hack external company during testingABC NewsTechnologyBBCTechnologyAug 6, 2026 · 1 minMeta says AI model accessed the internet and hacked another firmBBCTechnologyABC NewsTechnologyAug 6, 2026 · 1 minMajor change to NSW drivers licences to crack down on identity fraudABC NewsTechnologyThe GuardianTechnologyAug 6, 2026 · 1 minMeta says its AI model hacked into another company during testingThe Guardian
TechnologyABC NewsTechnologyAug 6, 2026 · 1 minMeta AI agent latest model to hack external company during testingABC News
TechnologyBBCTechnologyAug 6, 2026 · 1 minMeta says AI model accessed the internet and hacked another firmBBC
TechnologyABC NewsTechnologyAug 6, 2026 · 1 minMajor change to NSW drivers licences to crack down on identity fraudABC News
TechnologyThe GuardianTechnologyAug 6, 2026 · 1 minMeta says its AI model hacked into another company during testingThe Guardian