John Bohan announces Veo3.1, a model that generates photorealistic video from text prompts. It handles lighting, shadows, and physically plausible motion in a single pass, potentially replacing traditional camera crews and production pipelines.
Can you even tell what's real anymore? Because honestly, with VO3.1 Live Now, that question, it just got a whole lot harder to answer. What took cameras, entire crews, and weeks of back and forth, now it starts with an idea. A single prompt, and light behaves like light. Shadows fall exactly where they should. Movement that doesn't just look real, it feels real. VO3.1 actually gets it. The physics of how things move. The subtle motion that makes a moment believable. Even the emotional weight of a scene. This is VO 3.1, where the only limit left is what you can imagine.
