gl
o
signal
← All stories
Static
1 source
·
2h ago
Security researchers tricked LLMs into giving them cocaine recipes by abusing role models for prompt injection
If you want a picture of the future of LLM security, imagine Whac-a-Mole meets Groundhog Day
Related Stories
Vibe coding platform Base44 launches own model as AI startups seek defensibility
Sources: AI video startup Higgsfield is in talks to raise $300M to $500M at a $5B pre-money valuation, more than 4x its valuation in a January 2026 round
Meta Contractors Posed as Teens to Prompt Rival Chatbots About Suicide, Sex, and Drugs
Sources: hundreds of Meta contractors posed as minors to probe how competitor chatbots responded to prompts involving suicide, sex, and other high-risk subjects
'Djinn' Stealer Targets Cloud, AI Credentials