200,000 MCP servers expose a command execution flaw that Anthropic calls a feature

OX Security confirmed arbitrary command execution on six live platforms and estimates 200,000 MCP servers are exposed. Here's ...

Usage-based pricing killing your vibe - here's how to roll your own local AI coding agents

Do we even need Anthropic or OpenAI's top models, or can we get away with a smaller local model? Sure, it might be slower, ...
The Security Ledger

How Claude Planted Malicious Code In A Crypto-Trading App

A new report from ReversingLabs identified a new tactic by North Korean hackers: feeding malicious code to the AI systems ...
Hackaday

This Week In Security: State Malware, State Hardware Bans, And Stuxnet Before Stuxnet Was Cool

Making headlines everywhere is the CopyFail Linux kernel vulnerability, which allows local privilege escalation (LPE) from any user to root privileges on most kernels and distributions. Local ...

OpenAI Codex system prompt includes explicit directive to “never talk about goblins”

The system prompt for OpenAI’s Codex CLI contains a perplexing and repeated warning for the most recent GPT model to “never ...

Google issues new 'warning' in Microsoft Teams chat invitations and helpdesk scam

Cybercriminals are now leveraging Microsoft Teams, posing as IT helpdesk staff, to trick employees into revealing credentials ...

Why OpenAI's 'goblin' problem matters — and how you can release the goblins on your own

If OpenAI can accidentally train its flagship model to obsess over goblins, what other more subtle and potentially harmful ...