World Congress 2026 North America • Sep 26, 2026 • Session details

You Can’t Rerun Sunlight: Designing ML Data Architectures for Physical AI

An Phan

How do you debug physical AI when you cannot recompute reality? Learn why shifting to knowledge reproducibility prevents your autonomous agents from hallucinating false model flaws.

You Can’t Rerun Sunlight: Designing ML Data Architectures for Physical AI thumbnail

Checking access…

Playback and chapters load privately for Free videos.

Matching moments

1:26 min

Looking ahead at future limitations of AI models

Deivids Vilkinsons Deivids Vilkinsons +3 · World Congress 2026 Europe

2:22 min

Introduction to physical AI and the physical world

Moe Sani Moe Sani · World Congress 2026 Europe

3:04 min

Building trust and transparency into AI interactions

Ekaterina Streltsova Ekaterina Streltsova · Europe 2026 Virtual

1:51 min

Transitioning artificial intelligence from cloud environments to physical devices

Tomislav Tipurić Tomislav Tipurić +4 · World Congress 2026 North America

1:38 min

The tracing landscape for AI accountability

Amjad Shaikh Amjad Shaikh +1 · World Congress 2026 North America

1:20 min

Shifting focus from large language models to physical AI

Sergio Perez Sergio Perez · World Congress 2026 Europe