Baidu has unveiled Apollo World, an AI world model designed to simulate interactive road environments for training and testing robotaxis. The system aims to help autonomous vehicles learn from driving scenarios that are difficult to cover in the real world, including high-stakes events encountered on roads.
Across outlets, Baidu is described as using Apollo World to support its Apollo Go robotaxi efforts toward broader, global deployment. TechNode reports that Baidu says the simulator can generate large amounts of simulated driving each day, and that it produces real-time camera outputs as part of the training workflow. The South China Morning Post focuses on the simulator’s role in preparing robotaxis for rare but hazardous situations such as errant cyclists, animals entering roadways, and other unpredictable events.
The Express Tribune similarly frames the launch as part of Baidu’s plan to train Apollo Go’s robotaxi fleet for future expansion, emphasizing the model’s purpose rather than operational performance details. Overall, the coverage aligns on the product’s goal—better robotaxi training through simulated road environments—with differences mainly in which kinds of road hazards and implementation details each outlet highlights.
