Chips & SupplyUnited States
AMD to Acquire World Labs for $8.2B to Advance AI Models and Robotics
Useful
Analysts say the deal will help AMD inform future chip designs as AI evolves.
View the articleData Center Knowledge
Chips & SupplyUnited States
Analysts say the deal will help AMD inform future chip designs as AI evolves.
Analysts say the deal will help AMD inform future chip designs as AI evolves.
AMD has struck a deal to purchase AI research lab World Labs in an all-stock transaction valued at about $8.2 billion, the chipmaker announced Monday, Sept. 28. The deal aims to co-design hardware, software, and systems for emerging AI models and applications such as robotics and physical AI. The transaction is expected to close by the end of 2026, subject to customary closing conditions and regulatory approvals.
San Francisco-based World Labs develops spatial intelligence “world models” that use text, images, and video to generate and simulate interactive 3D environments, AMD said.
AI pioneer Fei-Fei Li, co-founder and CEO of World Labs, will join AMD as executive vice president and chief scientist. Upon closing, World Labs will continue to focus on AI model research, the company said.
“Building the compute platforms for the next generation of AI requires a deep understanding of how models are evolving,” said AMD CEO Lisa Su in a statement. “Fei-Fei and the World Labs team bring exceptional research leadership and model expertise. Together, we can use that insight to develop the hardware, software, and systems that will power the next generation of AI and strengthen the open AI ecosystem.”
Related: Core Scientific Doubles AI Capacity to 1.1 GW in $14B AMD Deal
Why the Deal Matters for AMD
Matt Kimball, vice president and principal analyst at Moor Insights & Strategy, said the acquisition allows AMD to understand chip requirements better as AI advances. AMD’s purchase doesn’t come out of left field because it had invested in World Labs previously, he noted.
“This isn’t about AMD buying a model company. This is about being able to understand the requirements of compute and memory – and the relationship between the two as the next generation of AI emerges,” Kimball told Data Center Knowledge in an interview. “I like the move. To me, it is AMD looking a few years out and capturing those needs today – needs that get baked into low-level architectural decisions and designs.”
“This isn’t about AMD buying a model company. This is about being able to understand the requirements of compute and memory – and the relationship between the two as the next generation of AI emerges." – Matt Kimball, vice president and principal analyst, Moor Insights & Strategy
Brendan Burke, a research director at Futurum Group, said the purchase puts AMD in a stronger position in the emerging physical AI market, which AMD estimates could reach $200 billion by 2035.
“AMD identified a transformative category that has yet to produce a significant commercial impact, and shows the company is using its recent momentum to think long-term about the future of AI,” Burke said. “You can’t hire a more accomplished researcher than Dr. Fei-Fei Li, and access to talent is the biggest advantage in AI training.”
Related: FLOPS vs Megawatts: Who’s Winning in 2026 Supercomputing?
Nvidia has already moved into physical AI with its Cosmos world model, along with its Isaac and Omniverse platforms and Project GR00T robotics models, which give it an early lead in the market, Kimball said.
Burke said the acquisition gives AMD access to leading researchers in the field of physical intelligence. They could help AMD train useful models and inform the design of AMD’s rack-scale systems for those model architectures, potentially helping AMD gain market share.
Burke added that AMD increasingly sees itself as an AI research organization, having invested in internal generative AI training efforts and creating open-source models. Buying World Labs allows AMD to move deeper into the market. Co-designing with World Labs could help AMD “move toward leadership in fields that aren't yet captured by leading labs,” he said.
Building World Models
World Labs builds world models, not large language models (LLMs). While LLMs focus on the relationship between language, tokens, and information, world models are about environments – geometry, physics, objects, motions, and dynamics, Kimball said.
Related: JPMorgan, OQC, and AMD Plan Quantum AI Data Center for Finance
“Think of it as a model for the physical world,” he said. “Any physical AI and robotics require all of this 3D modeling, geometric understanding, physics, etc., to work properly.”
For example, in training a robot to place a cup on a table, the model must account for the motions and contact dynamics so that the cup lands flat on the surface rather than half on and half off, he explained.
Talent and IP at the Core
Overall, World Labs is an acquisition of intellectual property, talent, and insights, Kimball said. The value of the deal extends beyond physical AI and into what comes next, he added.
“Fei-Fei is a rock star. You could argue that she is as much of the target as the company is. She and her team bring world-class talent to AMD, and this will enable the company to much more quickly find a competitive footing against Nvidia in this space,” he said.
About the Author
Wylie Wong is a journalist and freelance writer specializing in technology, business and sports. He previously worked at CNET, Computerworld and CRN and loves covering and learning about the advances and ever-changing dynamics of the technology industry. On the sports front, Wylie is co-author of Giants: Where Have You Gone, a where-are-they-now book on former San Francisco Giants. He previously launched and wrote a Giants blog for the San Jose Mercury News, and in recent years, has enjoyed writing about the intersection of technology and sports.
Want more Data Center Knowledge stories in your Google search results?
Editor's Choice
Industry Voices
Join the entire data center ecosystem September 21 – 23, 2026, at the Gaylord Texan Resort & Convention Center in Dallas, TX.