John O'Donnell

In this ten-minute lightning talk, Alice.io cuts through the hype to focus on the hardest problem in GenAI: trust, safety, and security at enterprise scale. Drawing on experience working with 8 of the world’s top 10 foundation model companies, we share real-world insight from where AI systems fail and how leading organizations prevent those failures before they reach production.

Alice.io helps Engineering, Cybersecurity, and Governance & Compliance teams at Fortune 500 companies stay ahead of PII and PHI exposure, fraud and abuse, prompt injection, data leakage, harmful and non-compliant content, and emerging adversarial threats as AI systems become more autonomous and embedded in critical workflows.

By extending protection beyond model-level guardrails with real-time prompt evaluation, red-teaming, and proactive threat detection, Alice.io enables enterprises to deploy resilient, responsible, and production-ready AI across regulated and high-risk environments.