Category: Security
-
One Line of Code. Eleven AI Models. All Bypassed.
•
Researchers at Trend Micro have documented a jailbreak technique called sockpuppeting — a method that bypasses the safety guardrails of 11 major AI models, including Claude, ChatGPT, and Gemini, using a single line of code. No hacking tools. No access to model internals. No sophisticated attack chain. Just one cleverly…
-
Is Your Phone Listening to You? The Answer Is More Unsettling Than You Think
•
A CBS Sunday Morning investigation asked the question millions of people have wondered about: is your smartphone secretly listening to your conversations to serve you targeted ads? Their answer, backed by two independent experts and a Northeastern University study of over 17,000 Android apps, was a firm no. No evidence…
-
One Hacker. Two AIs. Nine Government Agencies. Hundreds of Millions of Records.
•
Between late December 2025 and mid-February 2026, a single threat actor breached nine Mexican government organisations at federal, state, and municipal levels. The attacker exfiltrated data from databases containing hundreds of millions of citizen records, compromised hundreds of internal servers, built a live query API into government systems, and operationalised…