Loading posts ...

Skill2Exploit: When an Innocent Skill Makes Your Agent Deliver a Zero-Day RCE

Someone on your team asks their AI agent for a funny meme for the team chat. The agent uses a popular, harmless-looking skill, finds the image, and saves it to the company’s shared drive. Nothing…

What If a “Crash” Bug Could Take Over Your Nextcloud Server?

Getting a photo on Nextcloud (an open-source platform organizations host on their own servers to store, share and collaborate on files) is all fun and…
Read More

Why AI Incident Response Is Different When an AI Agent Is Compromised

Traditional incident response was built on well-understood security principles: users, endpoints, servers, applications, credentials, and network connections. AI agents introduce another category of principal: non-human…

The Muse House Rules: A Practical Guide to Securing Meta Muse

It’s 9 p.m. The office is empty, and most of your team logged off hours ago. But somewhere, an employee’s Meta Muse is still at…
Read More

The AI Enablement Standard: A Ready-to-Use RFP for AI Endpoint Security

Today, we’re announcing the AI enablement standard RFP, a new addition to Plutonium for security leaders. You can send it as is or customize it,…
Read More

NIST AI RMF: What It Actually Means for Enterprise Security Teams

Artificial intelligence (AI) risk management is rapidly becoming an enterprise security discipline. Yet many security teams still approach these systems using conventional application-security and third-party-risk…

Securing Grok Bot: A Hardening Guide to X’s Personal AI Assistant

Grok Bots are Cursor’s persistent personal AI assistants. A Bot works from a cloud computer shared with the user’s other Bots and, when allowed, reaches…
Read More