Storia in 8 fonti

AI Researchers Got Chatbots to Share Cocaine Recipes Using This One Wild Trick - Decrypt

Researchers say a new jailbreak technique tricked AI models into treating attacker-written text as their own reasoning, bypassing safety guardrails and exposing a deeper security flaw.

Raccontata da

Confronto fonti

6 prospettive sulla stessa storia

AI · summaries

decrypt.coStai leggendo8 h fa

AI Researchers Got Chatbots to Share Cocaine Recipes Using This One Wild Trick - Decrypt

Researchers say a new jailbreak technique tricked AI models into treating attacker-written text as their own reasoning, bypassing safety guardrails and exposing a deeper security flaw.

originale

Timeline cronologica

martedì 30 giugno 2026·theregister.com
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
martedì 30 giugno 2026·thehackernews.com
New BioShocking Attack Tricks AI Browsers Into Leaking User Credentials
LayerX says BioShocking used indirect prompt injection to trick six AI browsers into copying credentials from signed-in accounts.

AI Researchers Got Chatbots to Share Cocaine Recipes Using This One Wild Trick - Decrypt

Confronto fonti

AI Researchers Got Chatbots to Share Cocaine Recipes Using This One Wild Trick - Decrypt

Timeline cronologica

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

New BioShocking Attack Tricks AI Browsers Into Leaking User Credentials

AI researchers trick chatbots into sharing how to make cocaine as long as they believe a user is wearing a…

AI researchers bypass chatbot safety guardrails with new jailbreak technique called sockpuppeting

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

New BioShocking attack manipulates AI browser into data theft

New BioShocking Attack Tricks AI Browsers Into Leaking User Credentials

New BioShocking attack manipulates AI browser into data theft

GuardFall: When Decades-Old Shell Injection Tricks Beat Modern AI Safety Guardrails

AI researchers trick chatbots into sharing how to make cocaine as long as they believe a user is wearing a green shirt — 'CoT Forgery'…

'BioShocking' Attack Tricks AI Browsers Into Stealing Credentials

AI Researchers Got Chatbots to Share Cocaine Recipes Using This One Wild Trick - Decrypt

AI researchers bypass chatbot safety guardrails with new jailbreak technique called sockpuppeting