← all tags
// tag: jailbreak
fortnite's darth vader gave a researcher uranium enrichment instructions. epic called it a feature.
Epic's technical director accidentally said the truest thing anyone has said about AI safety guardrails — and it took Darth Vader to make them do it.
inject "sure, here is how to do it" and 11 ai models will finish the sentence
Trend Micro's sockpuppeting jailbreak works on 11 models — but the "immune" providers aren't safer, they just blocked a developer feature everyone else still ships.
researchers tricked AI agents into doing bad things by making them think they were playing BioShock
A new jailbreak PoC called 'BioShocking' wraps malicious instructions as game objectives — turns out agents are really into roleplaying.
AI models are jailbreaking each other with a 97% success rate
A peer-reviewed Nature Communications study found that reasoning models like DeepSeek-R1 and Grok 3 Mini can autonomously jailbreak other AI systems — no human required.
