Why video is the first frontier as AI learns to read the physical world

Lumana, founded by former Intel computer vision leaders, processes over a billion images daily from 50,000+ cameras using its VIA-1 model that learns normal patterns and flags anomalies locally before cloud transmission. With 562 million surveillance cameras outside China by 2025 and two-thirds shipped in 2024 having deep-learning analytics, video is a key area for physical AI leveraging existing infrastructure.