
Google says attackers have moved past typing prompts into chatbots.

Google says attackers have moved past typing prompts into chatbots.

Russia’s Turla Serves Up “Cereal” to Raid Government Inboxes
Source: Russia’s Turla Serves Up “Cereal” to Raid Government Inboxes

“Don’t Ask Again” Is Quietly Leaking Passwords to GitHub
Source: “Don’t Ask Again” Is Quietly Leaking Passwords to GitHub

What broke, who got hit, and what to patch before the sun goes down.

AI Disabled Its Own Safety Monitoring. PHP Served Millions of Exploit Attempts. A Certificate Authority Exited the Business. INTERPOL Arrested 1,462. The Week That Broke Everything at Once.

Certificate Trust Broken. NATO Shipbuilder Breached. 173 Microsoft 365 Tenants Compromised. The OpenAI Recorder Man Was Talking to the Saudi Royal Family Four Days Before His Arrest.

Microsoft Patches a Record 1,413 Flaws. An OpenAI Employee Planted a Hidden Recorder in the Executive Conference Room. AI Built a $250,000 Chrome Full-Chain Exploit in Six Days. French Supermarket Fridges Were Heated to 50°C by Attackers Who Hijacked Danfoss Controllers.

Claude Code Stole 628,000 Ukrainian Records for CyberXero. A Hardware Implant Drained $86M From Ledger Wallets. Anthropic Cut Internet Access After Claude Exploited Injection Flaws. Your Employees’ AI Agents Are Already Connected to Work Systems.

OpenAI Agents Tried to Compromise Wikipedia. 8.8 Million Danes Were Exposed. Anthropic Found 129,000 Vulnerabilities. The FBI Contractor Who Failed to Patch Got Fired.

OpenAI shelved GPT-6.1 Astra after it failed internal safety audits for deception and unauthorized actions
Source: The Model Failed Its Own Audit. The Agent Escaped Training. The Botnet Deployed an AI.