Waymo vehicles have reportedly logged more than 200 million miles of autonomous driving on public roads. But it has yet to hit a tornado or an elephant, and it is unlikely to respond well if it does. To try to help with those once-in-a-billion-mile scenarios, Waymo announced Friday that it’s introducing Waymo World Modela generative AI model that it will use to run in near-infinite situations to try to ensure that its vehicles are ready for the unpredictable, which also happens to fit the latest trends in the AI ​​space.

To be clear, Waymo’s global model is as important as any use case for the technology. The company has a ton of high-definition data collected from its time on the road that it can use to create realistic recreations of roads. But, the company said, instead of building a model based solely on that information, it will use Google’s Genie 3 model to put its cars in simulated situations beyond what is in the data set collected from cameras and lidar sensors.

We’re excited to introduce the Waymo World Model—a frontier generative mode for large-scale, hyper-realistic autonomous driving simulations built on @GoogleDeepMindThe Genie 3.

By simulating the “impossible”, we actively prepare Waymo Driver for some of the most unique and… pic.twitter.com/Pl80OMDqLC

— Waymo (@Waymo) February 6, 2026

Google made a splash last month when it released a beta version of Genie 3 to the publicwhich allows a subset of paid subscribers to create 3D worlds with realistic physics. Unlike a large language model (LLM)—the underlying technology that powers most AI tools including Google’s own Gemini—which uses large amounts of training data given to them to predict the most likely next part of a sequence, global models are trained on real-world dynamics, including physics and spatial properties, to create a simulation of how the physical environment behaves.

Genie 3 🤝 @Waymo

The Waymo World Model creates photorealistic, interactive environments to train autonomous vehicles.

It helps cars navigate through unusual, unpredictable events before actually encountering them. 🧵 pic.twitter.com/m6rlmkMFJH

— Google DeepMind (@GoogleDeepMind) February 6, 2026

Waymo plans to tap into that to put its cars through a gauntlet of scenarios they likely won’t see until it’s too late. That includes extreme weather conditions and natural disasters, so cars can learn how to navigate a tornado or floodwaters; sudden safety emergencies such as falling tree limbs or accidents with large amounts of waste; and run-in with the unexpected, like an elephant on the road. “By simulating the ‘impossible,’ we proactively prepare Waymo Driver for some of the most unique and complex scenarios,” the company said.

The theory is certainly sound, although the world models are not without their shortcomings. Early feedback on the consumer version of Genie 3 was a bit bleak, and the world’s models were prone to hallucinations. We are still in the very early stages of seeing these models deployed, and they have a lot of room for improvement.

And Waymos certainly has its issues in edge-case scenarios in the real world. Late last year, a Waymo ran into a beloved warehouse cat named Kit Katand last month, one encountered a child in a school zone. The interactions aren’t particularly unusual for a driver to find themselves in, so hopefully Waymo can refine its responses to those scenarios more than preparing for the most unlikely scenarios.