What Lies Beneath the Surface? Evaluating LLMs for Offensive Cyber Capabilities
Michael KouremetisMarissa DotterAlex ByrneDan MartinEthan MichalakGianpaolo RussoMichael Threet
Analysts investigate attack surfaces and defenses concerning What Lies Beneath the Surface? Evaluating LLMs for Offensive Cyber Capabilities.
Summary
Evaluating large language models for offensive cyber capabilities.
Related talks
The Linux Foundation·26 min
Divyansh Chaurasia
Reviews reducing Hallucinations and Evaluating LLMs for Production and evaluates key implications for security practitioners.
Black Hat USA·37 min
Richard Harang
Industry experts analyze emerging threats surrounding Practical LLM Security: Takeaways From a Year in the Trenches.
Black Hat USA·40 min
Bill Demirkapi
Engineers explore architecture and security mechanisms for Predict, Prioritize, Patch: How Microsoft Harnesses LLMs for Security Response.


