ended5월 20일· 1 sources
AgentThreatBench: The First OWASP Agentic Top 10 Security Benchmark
Why it matters
The AI safety community has a blind spot. We have excellent benchmarks for measuring whether an LLM will output harmful content (like toxicity or jailbreaks), and we have benchmarks for measuring whet...
1
Sources
+0
24h
—
Growth
124d
Active