Federated Learning Solves the Surgical Data Bottleneck
Lorenzo Sani explains why hospitals cannot share surgical video, making federated learning necessary for robotic surgery models.
Implementing Federated Learning to enhance Agentic AI
With Chris Benson, Lorenzo Sani · Friday, November 13, 2026
Chris Benson speaks with Lorenzo Sani, research scientist at Flower Labs and PhD researcher at Cambridge, about using federated learning to unlock private data for AI models. They examine how collaborative learning and privacy-preserving techniques enable training on complex agentic execution trajectories without leaking sensitive prompts or actions. The conversation also explores the operational benefits of open-weight models, high-stakes embodied AI in healthcare, and practical experiments with self-hosted agents at home.
Key takeaways
- Lorenzo Sani watched AI models command surgical robots firsthand, but noted that engineering teams still cannot explain why the model selected a specific physical step. Read more →
- Lorenzo Sani self-hosts open-weight language-vision models to run domestic agents without sending sensitive household execution trajectories to closed-source cloud APIs. Read more →
2 articles from this episode
Lorenzo Sani Runs Open-Weight Vision Models at Home for Privacy
Lorenzo Sani and Chris Benson explain how self-hosting open-weight models protects household vision data, from party games to wildlife cameras.
More Practical AI episodes
- Abundant Intelligence at the Midwest AI Summit Part 2Nov 8 · 4 articles
- Abundant Intelligence at the Midwest AI SummitNov 1 · 5 articles
- Exploit Development, Agentic Vulnerability Discovery, and Fake PoCsOct 25 · 6 articles
- Education is Failing our Students in this AI-Driven WorldOct 18 · 0 articles