ventali
Ventali Tan is the founder of Lita, where she leads the development of Valida, a high-performance zero-knowledge virtual machine. She studied physics, philosophy, and computer science at Duke University and previously worked in venture capital and AI in San Francisco. Before Lita, she co-founded Delendum, a cryptography research lab. Her current interests include designing systems with zero-knowledge proofs and exploring how AI may extend our understanding of natural sciences.
Session
Cryptography can’t tell us what is “good,” but it can provide verifiable ground truth about AI systems. Using tools like formal verification, zero-knowledge proofs, and verifiable computation, we can prove claims about what model ran, what guardrails executed, and how safety checks were enforced. This talk explores how cryptography can’t solve ethics, but can make AI behavior falsifiable, auditable, and accountable.