Hack the Future: Why Congress & White House Support AI Red Teaming
White House and Congressional leaders discuss government backing for the DEF CON Generative AI Red Team exercise to evaluate commercial LLM risks.
Summary
Government Collaboration with the Hacker Community
The White House, National Science Foundation, and Congressional AI Caucus formally partnered with DEF CON's AI Village to host the largest public Generative AI Red Team exercise.
Balancing Rapid AI Innovation with Emerging Risks
Austin Carson and federal leaders discuss the policy challenges of managing generative AI deployment across critical infrastructure, national security, and civil society.
Democratizing Public Trust and Safety Evaluations
Enabling thousands of independent hackers to stress-test frontier models from major AI providers creates transparent benchmarks for model hallucinations, bias, and vulnerability surfaces.
Institutionalizing Independent Red Teaming in AI Policy
The panel explores how lessons from public red-teaming exercises will inform future federal AI procurement standards and executive governance frameworks.
Related talks
Main Stage·47 min
Craig Martell
DoD Chief Digital and AI Officer Dr. Craig Martell examines the risks of military LLM adoption and the critical role of hackers in ensuring AI accountability.
Main Stage·43 min
Sven Cattell · Rumman Chowdhury · Austin Carson
AI Village organizers share the architecture and methodology behind the largest public Generative AI Red Team challenge testing frontier models at DEF CON 31.
Main Stage·45 min
Bruce Schneier
Bruce Schneier examines how artificial intelligence systems hack human institutions, tax codes, financial markets, and legal frameworks.