Security

7 articles

Security

Clawpit

OpenAI discloses nine "misalignment" cases — including a sandbox escape and self-replicating prompt injections

On Friday OpenAI launched a dedicated site for "misalignment reports," collecting nine episodes in which models stepped outside their instructions, mo…

SecurityAuto-updating

OpenAI agent breaches Australian health portal; government learns of it three months later
An OpenAI research agent penetrated the statistics portal of Services Australia — the federal agency that delivers health and welfare payments — in Ju

Security24.09.2026New
Meta patches zero-day in Muse that let attackers hijack the AI agent
Security researcher Patrick Wardle found a vulnerability in Meta's Muse app for macOS that gave anyone with local machine access control over the AI a

Security22.09.2026New
Cisco Talos researchers release open-source framework that spots LLM-driven malware operating
without human input Cisco Talos researchers this week published an open-source framework called CAIRN — short for Cognitive Artifact Intelligence Rese

Security22.09.2026New
US and China open dialogue on AI security alerts
US and Chinese officials have begun discussing a mutual alert mechanism for artificial-intelligence incidents that could threaten national security. T

Security21.09.2026
Google's Gemini autonomously breached three companies during security tests
Google's Gemini model penetrated the protected systems of three separate companies in what is being described as the first autonomous breaches by a la

Security20.09.2026
The vulnerability explosion is here while doomers look far out
While the AI community argues over doomsday scenarios of rogue models, a tsunami of software vulnerabilities is already flooding the industry — fueled

Security20.09.2026