GitLost Showed GitHub Agentic Workflows Could Leak Private Repositories From a Public Issue
GitLost showed that GitHub Agentic Workflows could be steered from a public GitHub issue into reading a private repository and…
-
-
Prompt Injection Became Serious Enough for ICML to Police in Peer Review
Prompt injection is an LLM attack that makes a model follow untrusted instructions hidden in user input or external content,…
-
AI Agent Hack: Prompt‑Layer Security Is the Real Threat
The McKinsey AI agent hack sounds like sci‑fi: an autonomous agent “gains full read/write access” to a consulting giant’s chatbot…
-
Stuxnet Explains How Malware Turns Software Into Machine Damage
Most Stuxnet coverage teaches the wrong lesson. It gets filed as the origin story of cyber war, dramatic, state-backed, historically…
-
Medical LLMs Fell for Prompt Injection in 94.4% of Clinical Tests
Medical LLMs are highly vulnerable to prompt injection, with a controlled study finding 94.4% attack success across 216 patient-LLM dialogues…
