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

Hundreds of millions of surveillance cameras already exist worldwide, but most footage still requires a human to search through it. Lumana, founded by ex-Intel computer vision leaders, is changing that by using its VIA-1 AI model to learn what is "normal" for each camera and flag unusual activity, processing over a billion images daily. By filtering footage locally before sending anything to the cloud, Lumana keeps costs manageable while reducing false alerts by up to 90%.