Demis Hassabis, CEO of Google’s AI research organization DeepMind, appeared to suggest Tuesday evening that Veo 3, Google’s latest video-generating model, could potentially be used for video games.
In response to a post on X beseeching Google to “Let me play a video game of my veo 3 videos already,” and asking, “playable world models wen?” Hassabis responded, “now wouldn’t that be something.”
On Wednesday morning, Logan Kilpatrick, lead product for Google’s AI Studio and Gemini API, chimed in with a reply: “🤐🤐🤐🤐”
Both posts from the Google executives are little more than playful suggestions, and a Google spokesperson told TechCrunch the company had nothing to share at the moment. But building playable world models isn’t outside the realm of possibilities for the tech giant.
World models are different from video-generation models. The former simulates the dynamics of a real-world environment, which lets agents predict how the world will evolve in response to their actions. Video-gen models synthesize realistic video sequences.
Google has plans to turn its multimodal foundation model, Gemini 2.5 Pro, into a world model that simulates aspects of the human brain. In December, DeepMind unveiled Genie 2, a model that can generate an “endless” variety of playable worlds. The following month, we reported that Google was forming a new team to work on AI models that can simulate the real world.
Others are working on building world models — most notably, AI pioneer Fei-Fei Li. Li came out of stealth last year with World Labs, a startup that has built its own AI system that generates video game-like, 3D scenes from a single image.
Veo 3, which is still in public preview, can create video as well as audio to go along with clips — anything from speech to soundtracks. While Veo 3 creates realistic movements by simulating real-world physics, it isn’t quite a world model yet. Instead, it could be used for cinematic storytelling in games, like cutscenes, trailers, and narrative prototyping
The model is also still a “passive output” generative model, and it (or a future Veo generation) would need to shift to a simulator that’s more active, interactive, and predictive.
But the real challenge with video game production isn’t just impressive visuals; it’s real-time, consistent, and controllable simulation. That’s why it might make sense to see Google take a hybrid approach that leverages Veo and Genie in the future, should it pursue video game or playable world development.
Google could find itself competing with Microsoft, Scenario, Runway, Pika, and, eventually, OpenAI’s video-generating model Sora.
Given Google’s planned moves in the world models space and its reputation for using its deep pockets and distribution muscle to steamroll rivals, competitors in this space would be wise to keep a close watch.
Source link