AMD agrees to buy Fei-Fei Li's World Labs for $8.2 billion in stock

AMD has agreed to pay about $8.2 billion in stock for World Labs, a San Francisco AI lab founded in 2024, and its announcement is plain about what the money buys: "a world-class team of researchers and model experts." Fei-Fei Li, World Labs' cofounder and chief executive, will become AMD's executive vice president and chief scientist, reporting to chief executive Lisa Su. The deal is expected to close by the end of 2026, subject to regulatory approvals and customary conditions, AMD said.

The leadership of the lab stays together. Justin Johnson and Ben Mildenhall, Li's cofounders, will work with her to continue leading the World Labs team as it joins AMD "to form a world leading frontier research organization," World Labs said in its own announcement. AMD said the group will keep its focus on AI model research after the close. Until then, AMD plans to keep World Labs separate from its chipmaking business, CNBC reported.

The price has two useful markers. It is AMD's second-largest acquisition on record, behind the roughly $50 billion it paid for Xilinx in 2022, according to CNBC. And seven months ago, AMD backed a $1 billion funding round in World Labs at a $5 billion valuation, SiliconANGLE reported, a round in which Nvidia also took part. AMD is now paying well above that mark to take full ownership of a company its chief rival had also invested in.

AMD's stated rationale is that World Labs' model expertise "will give AMD deeper insight into how workloads are evolving and help shape its future technology roadmaps." CNBC framed it more concretely: the lab's research will let AMD plan what its AI chips need to be capable of years in advance. In effect, AMD is bringing in a team whose output is foresight about its own customers' future demands.

Li put the case from the other side. "Without having a focused hardware effort, AI is hobbled in efficiency. And scale. And for our purposes, remains trapped in the digital world," she wrote in a post about the deal.

Neither side is a stranger to the other. Su was an early investor in World Labs, according to Li, and the two companies began a technical partnership last year on model training and inference optimization on AMD GPUs. Li appeared at AMD's CES presentation earlier this year, TechCrunch reported. AMD is acquiring people it has already worked alongside, not an unknown lab.

The skills involved are not the ones behind chatbots. World models are trained on data such as video depicting the physical, three-dimensional world rather than on text, Fortune reported, with the aim of AI that understands how objects move, a capability many in the industry see as crucial for humanoid robots and self-driving cars. Li wrote that the work requires expertise spanning "data, computer vision, large model training, systems and even computer graphics," and that she had assembled a founding team with world experts in each. World Labs also bought SceniX to build robotics simulation capability, and its products include Marble, which creates 3D worlds from prompts, and a model called Atlas.

Nvidia already offers open-weight world models such as Cosmos, while AMD has offered the public only text- and video-based models, TechCrunch reported. Nvidia has also partnered with startups such as Wayve on world models for self-driving cars, according to Fortune, and Yann LeCun, Meta's former chief AI scientist, launched a rival world-model startup, AMI Labs, in March.

The deal lands in a market where large technology companies are paying heavily for top engineering and research talent, even in young businesses, CNBC reported. OpenAI paid about $6.4 billion last year for io, the AI devices startup of former iPhone designer Jony Ive. Meta spent $14 billion in 2025 for a minority stake in Scale AI in a deal that brought founder Alexandr Wang in to start a new AI division. In each case, the purchase was built around specific people.

AMD shares were roughly flat in after-hours trading Monday, Fortune reported. Li, for her part, cast the move as the destination rather than an exit: "My career has led up to this moment - to continue growing and running a frontier research team, within a leading AI company."