The $8.2 Billion Payday for the 'Godmother of AI'
Published October 18, 2026
AMD's $8.2 billion all-stock acquisition of World Labs, announced September 28, 2026, will lift co-founder Fei-Fei Li's net worth to at least $1.6 billion when it closes, per the Bloomberg Billionaires Index. The Stanford professor who built the ImageNet dataset that launched modern computer vision became a billionaire earlier this year and now joins AMD as executive vice president and chief scientist.
On September 28, 2026, chipmaker AMD announced it would acquire World Labs, a two-year-old artificial intelligence startup, in an all-stock deal valued at roughly $8.2 billion. For the startup's co-founder, Stanford professor Fei-Fei Li, the agreement was a second kind of milestone: the Bloomberg Billionaires Index calculated that her net worth would rise to at least $1.6 billion once the transaction closes, an increase of more than half from her already considerable wealth. It is a rare thing for a career academic to cross the ten-figure line at all; Li has now done it twice in a single year.
Li's reputation was built long before she founded a company. In 2009 she created ImageNet, the massive labeled image dataset that became the training ground for the deep-learning breakthroughs in computer vision — the work that earned her the nickname 'the godmother of AI.' She later directed Stanford's AI lab. In January 2024, she co-founded World Labs with researchers Justin Johnson, Ben Mildenhall and Christoph Lassner to pursue what she called spatial intelligence: AI systems that understand and simulate physical, three-dimensional environments rather than just processing text.
The bet drew serious money fast. World Labs raised roughly $1.2 billion across two funding rounds, backed by names including a16z, Nvidia, Autodesk, Fidelity and — notably — AMD itself, which joined a $1 billion round in February 2026 that valued the company at more than $5 billion. That round is what first made Li a billionaire. By November 2025 the lab had shipped its first commercial product, Marble, which turns a photo or text prompt into an editable 3D scene; in September 2026 it released Atlas, a more advanced world model aimed at robot training.
The AMD deal, announced by the chipmaker and widely reported by Bloomberg, CNBC and the Wall Street Journal, is AMD's second-largest acquisition ever, trailing only its roughly $49 billion purchase of Xilinx in 2022. It is an all-stock transaction expected to close by the end of 2026, pending regulatory approval, and it comes with a job: Li will join AMD as executive vice president and chief scientist, reporting directly to chair and CEO Lisa Su. AMD's stated goal is to understand where AI models are headed so it can build the hardware those models will need — and to close ground on rival Nvidia in the race for world-model technology.
What makes the story remarkable is the compression of the timeline. Li went from respected professor to billionaire to the centerpiece of an $8.2 billion acquisition in under three years. Her co-founders and early employees stand to share in the windfall through the all-stock structure. And the arc says something durable about where money in technology now flows: the most valuable currency in the AI boom is not necessarily a product — it is a team that can see around the corner of the technology, and a chip giant willing to pay $8.2 billion to hire one.
The lesson: Foundational research compounds: a career building the building blocks of AI became a $1.6 billion stake.
Source: original reporting · How we source stories