AMD Acquires World Labs for $8.2 Billion to Advance Physical AI

Advanced Micro Devices (AMD) has agreed to acquire artificial intelligence startup World Labs in an all-stock transaction valued at approximately $8.2 billion. The deal aims to integrate frontier research into the AMD hardware roadmap to advance physical AI, and the companies expect the transaction to close by the end of 2026.

Following the acquisition, World Labs founder Fei-Fei Li will join AMD as executive vice president and chief scientist. She will report directly to CEO Lisa Su. This move marks a strategic change for AMD as it competes with Nvidia by embedding specialized AI capabilities into its silicon and software ecosystems.

"The more you understand end to end, the better system you are going to build. That’s why we are acquiring World Labs. To have access to the world-class talent Fei-Fei has brought together and bring it together with AMD’s capabilities around hardware, software, and systems," said AMD CEO Lisa Su, according to a LinkedIn report.

Defining 'Physical AI'

World Labs focuses on physical AI, a branch of machine learning that teaches intelligent agents how to interact with the physical world by simulating environments. This research allows AI models to learn within physics-aware digital simulations before deployment into real-world applications like robotics or autonomous vehicles.

"Intelligent agents, whether it's robots or vehicles or even tools, can learn inside very rich physics-aware digital worlds before they even need to be deployed into the real one, making them much safer," said Fei-Fei Li, founder of World Labs, as cited by CNBC.

Integration and Market Strategy

AMD plans to operate World Labs as a separate entity from its core chipmaking business until the transaction officially closes. By bringing World Labs research capabilities in-house, AMD intends to offer a vertically integrated stack that combines its GPU and NPU architectures with proprietary AI models for spatial intelligence and physical interaction.

This acquisition arrives as the semiconductor industry faces pressure to demonstrate how hardware supports the next generation of AI agents. While Nvidia maintains a position in data center AI training, AMD is betting on physical AI to capture the market for edge-based robotics and autonomous systems, where hardware-software co-design helps reduce latency and improve safety.