Search papers, labs, and topics across Lattice.
University of California
3
0
4
VIGIL achieves over 95% recall in detecting policy violations in AI agent skills, significantly enhancing runtime enforcement capabilities.
Unleashing AI agents to find zero-day exploits requires more than just better models: AgentFlow's automated harness synthesis just discovered 10 new Chrome vulnerabilities, including two critical sandbox escapes.
Your LLM agent's security is only as strong as the cheapest API router it uses: researchers found malicious code injection and credential theft rampant in both paid and free LLM routers.
