AMD acquires World Labs for $8.2 billion to shape AI hardware
According to announcements on September 28, 2026, AMD agreed to acquire World Labs, an AI research lab founded by Fei-Fei Li, for $8.2 billion in an all-stock transaction. Fei-Fei Li will join AMD as executive vice president and chief scientist. The deal is expected to close by year-end subject to regulatory approval. World Labs develops spatial-intelligence models that generate and simulate interactive 3D environments from text, images and video, technology also applied to robotic learning and simulation. This acquisition ranks as AMD's second-largest ever, after its approximately $50 billion Xilinx purchase in 2022.
In short
- AMD's $8.2 billion acquisition of World Labs marks a shift from chip-only strategy to vertically integrated AI infrastructure combining hardware, models and research.
- The deal gives AMD direct access to emerging physical AI and robotics workload patterns, informing chip design for next-generation applications.
- Fei-Fei Li, creator of ImageNet, joins AMD as Chief Scientist, signalling AMD's commitment to AI research leadership as competition with Nvidia intensifies.
- Large technology companies are consolidating AI talent and capability through acquisitions in a race to build complete AI stacks.
- The transaction is AMD's second-largest ever acquisition; closing expected by year-end 2026 subject to regulatory approval.
What AMD is acquiring
On September 28, 2026, chipmaker Advanced Micro Devices announced a definitive agreement to acquire World Labs, an AI research lab founded by AI pioneer Fei-Fei Li, for $8.2 billion. The transaction is all-stock and expected to close by the end of 2026, subject to regulatory approval.
World Labs develops spatial-intelligence models that generate, reconstruct and simulate interactive 3D environments from text, images and video inputs. The company also builds technology for robotic learning and simulation. Its flagship product, Marble, shipped commercially in November 2025 and creates navigable 3D environments from single prompts, images or video clips.
Why the hardware giant is buying AI research
AMD has publicly stated that the acquisition addresses a strategic gap: understanding how next-generation AI workloads actually behave. Unlike pure consumer AI models, spatial intelligence and physical AI impose different computational demands on processors. By bringing World Labs' research team inside AMD, the company gains direct insight into how these workloads evolve, which informs chip design years in advance.
According to AMD's announcement, AI is expanding beyond traditional generative AI into reasoning, robotics, simulation and physical AI, each creating distinct infrastructure requirements. World Labs' model-development expertise gives AMD visibility into these emerging patterns, helping shape future hardware roadmaps. AMD CEO Lisa Su stated the motivation plainly: "The more you understand end to end, the better system you are going to build."
This acquisition also positions AMD against Nvidia in the AI infrastructure market. Nvidia has already expanded beyond chips into software, models and services; it recently agreed to acquire the Hugging Face open-source platform for approximately $13 billion and offers its own Cosmos family of world models.
Fei-Fei Li's role and track record
Fei-Fei Li, a Stanford computer science professor who previously led AI research at Google, will join AMD as executive vice president and chief scientist, reporting directly to CEO Lisa Su. Li is widely recognised for pioneering work in AI, particularly for creating the ImageNet database two decades ago, which sparked a wave of AI advancement in computer vision and inspired decades of research.
AMD and World Labs already had existing ties before the acquisition. The two companies formed an inference optimization-and-training partnership in 2025, and Li attended AMD's CES 2026 presentation in January, signalling the relationship had already deepened.
Scale of the transaction
The $8.2 billion all-stock deal represents AMD's second-largest acquisition on record. The company's only larger deal was its roughly $50 billion acquisition of semiconductor company Xilinx, completed in 2022. The transaction reflects the intensity of competition among major technology companies to acquire AI research talent and capabilities.
According to Citigroup analyst Atif Malik, the deal serves three key purposes for AMD: acquiring top AI engineering talent, positioning the company within physical AI, and building end-to-end systems that span hardware, software and models. The analyst noted that the transaction strengthens AMD's competitive position against Nvidia and accelerates its broader AI platform ambitions.
Industry context: the broader AI acquisition race
AMD's purchase of World Labs reflects a wider trend among technology giants deploying capital to secure AI research capacity and talent. OpenAI acquired Jony Ive's AI devices startup io for approximately $6.4 billion in 2025. Nvidia spent a combined $33 billion on AI chipmaker Groq and agreed to acquire Hugging Face for approximately $13 billion. Meta invested $14 billion in 2025 for a minority stake in Scale AI, bringing in founder Alexandr Wang to lead a new AI division.
These acquisitions signal that tech companies no longer view chips or models in isolation. Instead, they are building vertically integrated stacks that combine silicon, software, research capacity and operational systems. For businesses investing in automation infrastructure, this consolidation may affect the diversity of suppliers and the ease of switching between providers.
What to watch
The deal closes by year-end 2026 pending regulatory approval. Once complete, the World Labs team will continue focusing on AI model research under AMD's organisational structure. The pace at which AMD integrates World Labs' research into its chip roadmap will be a practical indicator of whether this investment yields hardware advantages. Industry observers will also track whether physical AI models and robotic simulation workloads grow fast enough to justify an $8.2 billion investment today.
How AiStaffo would automate this
While this deal primarily concerns chip manufacturers and large AI labs, it has downstream effects on businesses automating operations. As AMD integrates World Labs' spatial-intelligence models, three practical implications emerge. First, automation platforms built on AMD infrastructure will gain access to more advanced 3D environment understanding, enabling robots and physical systems to learn faster in simulation before real-world deployment. Second, the consolidation of AI capability within fewer large suppliers may reduce choice in which models and hardware small and mid-size businesses can access for back-office automation. Third, AMD's new research team will prioritise use cases that drive expensive GPU demand, potentially making cutting-edge models available faster to automation builders but at higher computational cost. For businesses automating data handling, document processing and routine workflows that don't require physical AI or robotics, this acquisition changes little in the near term. For those building warehouse automation, robotic process handling or physical inventory systems, the timeline to advanced simulation-trained agents may accelerate. Book a free automation audit to understand how new AI capabilities affect your specific automation roadmap.
Questions people ask
Why is AMD buying a research lab instead of building AI models itself?
What does World Labs actually do?
Who is Fei-Fei Li and why does she matter?
When will this deal close?
How does this affect the competition between AMD and Nvidia?
Book a free automation audit
Thirty minutes. We look at one process you run every week and tell you exactly what an AI worker would take off your desk, and what it would not.
