Back to AIBriefs
AnalysisCybersecurityAI Agents

MosaicLeaks: New benchmark tests if research agents keep secrets

MosaicLeaks is a new benchmark for measuring whether AI research agents inadvertently leak sensitive information. The assessment includes tasks requiring agents to work with confidential data while maintaining secrecy.

2 hours ago
MosaicLeaks: New benchmark tests if research agents keep secrets — AIBriefs