Victoria Westerhoff

Science-driven Strategy

What do neuroscience, strategy, national security, and AI red teaming have in common?

Decision Making

(and me)

Podcasts

Microsoft Research Podcast

Ahead of the Breach Podcast

Cyb3rMaddy Interview

Talk Python to Me Podcast

Copilot L33t Spe4k Podcast

Articles

Fast Co. This Microsoft Security Team Stress Tests AI for Worst-Case Scenarios

-El Pais: Neurocientíficos, militares y hasta un preso: así trabaja el equipo que ‘hackea’ la IA de Microsoft antes de que llegue al público

-Blick: Tori Westerhoff Talks About the Future of AI Security

- In-House Evaluation Is Not Enough

- Lessons Learned from Red Teaming 100 GenAI Products

- Phi-3 Safety Post Training: BreakFix Cycles

- PyRIT: A Framework for Security Risk Identification and Red Teaming

- National Defense: Strengthening AI Models by Attacking Them

-Axios: Microsoft injects AI agents into security tools

Conferences

- Future of 3rd Party Evaluation Workshop 2024

- Black Hat USA 2025: AI Summit

- CSET 2024 What's Next for Red Teaming

- Build 2024: How Microsoft Does AI Red Teaming