Secure collaborative dataset generation for privacy-preserving AI

We propose to design and build the first-ever decentralized platform that enables the creation of training datasets in a privacy-preserving manner to power the next generation of domain-specific AI models.

Biography

How do we build AI that learns from sensitive data without ever exposing it? John (Ioannis) Liagouris is an Assistant Professor of Computer Science at Boston University, where he co-leads the CASP Lab (Complex Analytics and Scalable Processing). His research builds data processing systems with strong cryptographic privacy guarantees, using secure multi-party computation (MPC) and differential privacy to let organizations collaborate on sensitive datasets without surrendering control. Previously a senior researcher at ETH Zurich and visiting scholar at UC Berkeley.