Large-scale capital deployment into AI-focused hardware, cloud capacity and data-center projects is accelerating as corporations and AI developers sign a string of high-value, multi-year arrangements. The deals range from chip supply and equity options to long-term cloud contracts and data-center construction ventures, collectively representing hundreds of billions of dollars in committed or potential spending.
Below is a sector-by-sector account of the recent agreements reported between chipmakers, cloud providers, AI developers and data-center investors. The listings preserve the contractual contours as disclosed, including purchase volumes, investment ranges and any equity options noted in the announcements or reports.
OpenAI-related arrangements
Several of the largest reported commitments center on OpenAI, the AI developer behind ChatGPT. The reported agreements include both direct investments and large-scale sourcing deals for compute and processor hardware.
- Amazon and OpenAI - Amazon has been described as considering an investment in OpenAI on the order of $10 billion, but parties have characterized the discussions as "very fluid."
- Disney and OpenAI - Walt Disney is reported to be investing $1 billion in OpenAI and licensing its characters from franchises such as Star Wars, Pixar and Marvel for use in OpenAI’s Sora AI video generator. Under a three-year licensing arrangement, Sora and ChatGPT Images are to begin producing videos featuring licensed Disney characters early next year, though the agreement explicitly excludes talent likenesses and voices.
- Broadcom and OpenAI - OpenAI has partnered with Broadcom to develop its first in-house AI processors, a move that aligns with the startup’s drive to secure additional computing horsepower through custom silicon.
- AMD and OpenAI - Advanced Micro Devices agreed to supply AI chips to OpenAI under a multi-year arrangement that also grants OpenAI the option to acquire up to roughly 10% of AMD.
- Nvidia and OpenAI - Nvidia is reported to be planning an up-to-$100 billion investment in OpenAI and will supply data-center chips, with the deal giving Nvidia a financial stake in OpenAI. OpenAI is already described as an important Nvidia customer.
- Oracle and OpenAI - Oracle is reported to have agreed to a large cloud-computing arrangement with OpenAI under which the latter is expected to purchase $300 billion in computing power over about five years.
- CoreWeave and OpenAI - CoreWeave signed a five-year contract with OpenAI valued at $11.9 billion in March, completing the agreement prior to CoreWeave’s Nvidia-backed initial public offering.
- Stargate datacenter project - A joint venture involving SoftBank, OpenAI and Oracle to build data centers was announced, with public commentary noting plans to invest up to $500 billion to fund infrastructure for artificial intelligence. The project was highlighted as a large-scale construction and capacity initiative.
Meta Platforms - securing chips, cloud and AI talent
Meta Platforms has been active across several fronts as it scales AI compute and seeks to integrate advanced AI into its consumer products.
- Meta and AMD - AMD agreed to sell up to $60 billion worth of AI chips to Meta over five years. The agreement also allows Meta to purchase up to a 10% stake in AMD. As part of the supply plan, AMD will provide six gigawatts of chips to Meta, beginning with one gigawatt of the yet-to-be-released MI450 hardware in the second half of this year. Meta’s planned purchases extend beyond flagship graphics processors to include central processors, including a customized variant tailored for Meta’s requirements.
- Meta and Manus - Meta is reported to be acquiring the Chinese startup Manus as part of efforts to embed more agentic AI capabilities into platforms including Facebook, Instagram and WhatsApp. The transaction’s valuation is reported by a source to be between $2 billion and $3 billion; explicit financial terms were not disclosed.
- Meta and CoreWeave - CoreWeave has agreed to supply Meta with computing power under a $14 billion arrangement.
- Meta and Oracle - Oracle is reported to be in talks with Meta for a multi-year cloud computing agreement worth approximately $20 billion, a deal framed as part of Meta’s strategy to secure faster cloud access.
- Meta and Google - Google previously signed a six-year cloud computing agreement with Meta worth in excess of $10 billion.
- Meta and Scale AI - Meta acquired a 49% stake in Scale AI for roughly $14.3 billion and installed Scale’s 28-year-old CEO, Alexandr Wang, in a prominent role within Meta’s AI strategy.
Nvidia and related hardware strategies
Nvidia features both as a supplier and investor in multiple arrangements, reflecting its central role in AI compute supply chains.
- Nvidia and Groq - Nvidia has agreed to license chip technology from Groq and hire Groq’s CEO Jonathan Ross along with other engineers. Reports indicate Nvidia agreed to acquire Groq’s assets for $20 billion.
- Microsoft, Nvidia and Anthropic - Microsoft will invest up to $5 billion and Nvidia up to $10 billion in Anthropic. Under the arrangement, Anthropic will commit $30 billion in running its workloads on Microsoft’s cloud, and pledge up to 1 gigawatt of compute powered by Nvidia’s Grace Blackwell and Vera Rubin hardware. The parties are also set to collaborate on improving chips and models for higher performance.
- Nvidia-backed investor group and Aligned Data Centers - An investor consortium that includes BlackRock, Microsoft and Nvidia is acquiring U.S.-based Aligned Data Centers, which operates nearly 80 facilities, in a deal valued at $40 billion.
- Nvidia and Intel - Nvidia is set to invest $5 billion in Intel, an arrangement that would result in Nvidia receiving roughly 4% of Intel after issuance of new shares.
- CoreWeave and Nvidia - CoreWeave entered an initial order valued at $6.3 billion with Nvidia; part of the agreement guarantees Nvidia will purchase any cloud capacity that CoreWeave does not sell to customers.
Google's data-center and software investments
- Google and Texas - Google will invest $40 billion to build three new data centers in Texas through 2027, including facilities in Armstrong County and Haskell County, alongside continued investment in its existing Midlothian campus and Dallas cloud region.
- Google and Windsurf - Google hired several key employees from AI code-generation startup Windsurf and will pay $2.4 billion in license fees under a non-exclusive agreement to use certain Windsurf technologies.
Other notable infrastructure and chip deals
- Nebius Group and Microsoft - Nebius Group will provide Microsoft with GPU infrastructure capacity under a deal valued at $17.4 billion over five years.
- Intel and SoftBank Group - SoftBank Group is injecting $2 billion of capital into Intel, a move that makes the Japanese investor one of Intel’s top-10 shareholders.
- Tesla and Samsung - Tesla signed a $16.5 billion agreement with Samsung Electronics to source chips, with Tesla’s CEO stating the South Korean firm’s new Texas factory will produce Tesla’s forthcoming AI6 chip.
- Amazon and Anthropic - Amazon invested $4 billion in Anthropic, effectively doubling its prior investment in the company behind the Claude chatbot.
- SoftBank and DigitalBridge - SoftBank Group will acquire digital infrastructure investor DigitalBridge Group in a transaction valued at $4 billion.
What these deals collectively indicate
The agreements reported above span raw silicon supply, custom processor development, long-term cloud compute purchase commitments and data-center construction. They highlight two clear themes: first, that AI developers and platform companies are seeking long-term capacity and preferential supply arrangements for compute; and second, that chip vendors and data-center owners are embedding themselves into AI ecosystems through both commercial contracts and equity or asset transactions.
While the financial scope of these arrangements varies - from single-digit billion-dollar deals to commitments measured in the hundreds of billions - the common thread is multi-year planning and a push to secure scale, whether through capital investments, licensing or long-term purchase agreements.
Several agreements include equity options or investment components, effectively aligning suppliers and buyers through ownership stakes as well as commercial terms. Other arrangements lock in large volumes of capacity over multi-year horizons, reflecting the projected ongoing compute needs of advanced AI model training and deployment.