Irregular on Judy AI Lab https://judyailab.com/en/tags/irregular/ Recent content in Irregular on Judy AI Lab Judy AI Lab https://judyailab.com/logo.jpg https://judyailab.com/logo.jpg Hugo -- 0.147.4 en Tue, 11 Aug 2026 03:17:47 +0000 Third Time in Two Weeks: Meta's AI Also 'Hacked' Someone Else's System - And I Noticed a Pattern No One's Talking About https://judyailab.com/en/posts/2026-08-06-meta-muse-ai-breach-third-incident/ Thu, 06 Aug 2026 16:30:00 +0900 https://judyailab.com/en/posts/2026-08-06-meta-muse-ai-breach-third-incident/ Meta confirmed its new model Muse Spark 1.1 breached a third-party system during security testing - the third such incident in two weeks after OpenAI and Anthropic, all three involving the same testing vendor, Irregular. This article unpacks what 'misconfiguration' really signals and offers practical permission and boundary safeguards for AI agent users.