Fetching latest headlines…
Security researchers tricked LLMs into giving them cocaine recipes by abusing role models for prompt injection
EUROPE
🇬🇧 United Kingdom•June 29, 2026

Security researchers tricked LLMs into giving them cocaine recipes by abusing role models for prompt injection

3 views0 likes0 comments
Originally published byThe Register
If you want a picture of the future of LLM security, imagine Whac-a-Mole meets Groundhog Day

Comments (0)

Sign in to join the discussion

Be the first to comment!

🇬🇧

United Kingdom

EUROPE

More news from United Kingdom

Related News

Finland’s CurifyLabs raises €12 million to expand 3D printing-based medicine compounding

Finland’s CurifyLabs raises €12 million to expand 3D printing-based medicine compounding

22h ago

NASA says it will isolate volunteers from the outside world for a year

NASA says it will isolate volunteers from the outside world for a year

3d ago

NetNut cracked as Google and FBI target 2 million-device botnet

NetNut cracked as Google and FBI target 2 million-device botnet

3d ago

Google DeepMind and A24 announce first-of-its-kind research partnership

Google DeepMind and A24 announce first-of-its-kind research partnership

3d ago

London’s geoSurge raises €10 million to help brands understand AI-generated outputs

London’s geoSurge raises €10 million to help brands understand AI-generated outputs

3d ago

View all EUROPE news →