AMD Signs Multibillion-Dollar Chip and Investment Deal With Anthropic
AMD will invest up to $5 billion in Anthropic as the AI company agrees to buy 2 gigawatts of AMD's next-generation Instinct MI450 chips starting in 2027.

AMD just landed one of the biggest wins yet in its push to challenge Nvidia.
The chipmaker signed a deal with Anthropic worth tens of billions of dollars in AI servers while agreeing to invest up to $5 billion directly in the AI company, according to the Wall Street Journal.
The agreement strengthens AMD’s foothold in a chip market still dominated by Nvidia, which already predicts a trillion-dollar AI outlook.
Reports also noted Anthropic will deploy up to 2 gigawatts of AMD’s newest chips, with the first gigawatt coming online in the first half of next year.
Inside the Chips-for-Equity Arrangement
CNBC reports Anthropic will deploy up to 2 gigawatts of AMD’s Instinct MI450 Series GPUs in AMD’s Helios rack-scale systems, with the first gigawatt coming online in the first half of next year, a scale CNBC says has become a standard measure of AI data center power.
AMD’s investment of up to $5 billion is its first with Anthropic, with funding tied to deployment milestones.
AMD CEO Lisa Su said the companies have been working together for some time and that AMD wants to play a major role in Anthropic’s infrastructure.
Anthropic currently uses Google’s tensor processing units, Amazon’s Trainium chips, and Nvidia GPUs, and will add AMD chips to its own data centers while leasing additional capacity from cloud providers and AI-focused cloud operators.
Why This Matters for Both Companies
Reuters describes the agreement as the latest example of a circular AI deal, where chipmakers invest in AI companies that are also major customers.
This strategy is mirrored in Nvidia’s reported talks to invest up to $30 billion in OpenAI, while also providing chips for the company’s flagship models.
Emarketer analyst Jacob Bourne told Reuters the deal is strategically significant because it strengthens AMD’s position as the number two player behind Nvidia.
Notably, AMD has also signed major deals with Meta and OpenAI as well, and is now competing for younger AI startups.
Su said AMD also signed an engineering partnership with Anthropic, using Claude models to improve its chip technology and launch flagship AI processors.
However, AMD’s chips still trail Nvidia’s in demand, partly because AI companies are less familiar with its software, leading AMD to offer incentives such as warrants for future shares.
Anthropic’s Growing Computing Appetite
For the Claude creator, the deal addresses a real capacity crunch. The surging demand for its AI tools has previously forced the company to limit usage, with some customers reporting frequent outages.
Earlier this year, Anthropic signed additional infrastructure deals with Amazon and Google, even as the search giant continues to lose top mindshare to rival AI labs.
Su noted that securing gigawatt-scale computing capacity isn’t something companies can arrange overnight, explaining that such deployments require planning 12, 18, or even 24 months in advance.
AMD is also in reported talks to provide a financial backstop for Anthropic’s future data-center leases, a role that mirrors how Google has already agreed to backstop some of Anthropic’s data-center deals to help secure access to its TPUs.
And reportedly, MoonShot AI has unveiled Kimi K3 as well to have a standard rivalry with Anthropic.
Source: Exclusive | AMD and Anthropic Sign Major Chips-and-Investment Deal



