Researchers warn US politics is repeating its ChatGPT mistake with world models

Researchers say US policymakers are underestimating world models and physical AI, repeating the slow response seen with large language models.

World models use multimodal sensor data to predict physical environments and could reshape robotics, drug discovery, driving, surveillance, labor, and national security.

Experts warn China is moving faster in robotics while the US lags, creating a growing strategic gap in advanced AI.

Compute, hardware, and supply-chain readiness are becoming major bottlenecks, making policy delay increasingly costly for the US.