Cisco Talos tracks UAT-11985, which uses AI-written event lures and an adversary-in-the-middle kit to steal Google logins from Taiwan researchers
Adversa AI says encrypted prompt injections on web pages can trick GitHub Copilot CLI into leaking .env secrets. GitHub declined
Google stopped accepting new submissions to its open-source bug bounty program after a flood of low-quality, AI-generated vulnerability reports arrived.
The FTC confirmed it is investigating OpenAI, Anthropic, and other AI firms after agents reportedly went beyond instructions to hack
A 16-year-old researcher used a self-built AI tool to gain admin access to Microsoft’s internal Titan analytics platform, exposing 17.3
An autonomous AI agent exploited a vulnerability to breach Dutch nonprofit DIVD on its own, leaving extensive forensic evidence of
OpenAI disclosed that GPT models can be manipulated into self-replicating prompt injections that spread like a worm across connected tools
Glow Security found over 13,000 sensitive screenshots from AI coding agents publicly exposed on GitHub across 343 companies in a
OpenAI canceled GPT-6.1 Astra release after agents bypassed access controls, attacked Australian government sites, and failed alignment testing.
JadePuffer ransomware group used autonomous AI agents to delete Azure virtual machines, databases, and storage after hijacking service principals.
Subscribe to the Daily Security Review Email Briefing to stay informed on the latest threats, trends, and technology, along with insightful columns from industry experts.