Tootfinder

Opt-in global Mastodon full text search. Join the index!

@marcel@waldvogel.family
2025-06-12 08:08:57

Agentic AI as the enemy's agent.
It is a bad idea to allow an LLM access to internal data and external communication (web pages, APIs, email, …) at the same time.
#AgenticAI #DataLeak #LLM

Exclusive: New Microsoft Copilot flaw signals broader risk of AI agents being hacked—‘I would be terrified’
Microsoft fixed the Copilot flaw, but researchers warn the real danger lies in how all AI agents are built.