Intel has officially announced that it is joining the Terafab AI chip fab park initiative under Elon Musk, teaming up with SpaceX, xAI, and Tesla to build a chip production ecosystem capable of delivering up to 1 terawatt of compute power per year.
(Background: Musk wants to build the world’s largest chip factory! TeraFab targets producing 200 billion chips per year to surpass TSMC.)
(Additional context: SoftBank partners with Intel to break the AI chip deadlock! It’s rumored that Intel’s technology isn’t up to the task, so it’s looking for TSMC instead, while pushing back against NVIDIA.)
Table of Contents
Toggle
On the 7th, Intel (intel) announced that it has officially joined the Terafab AI chip park initiative under Musk, and will work together with SpaceX, xAI, and Tesla to jointly manufacture processors, driving Musk’s ambitious plans in the robotics and data center space.
Terafab is not a typical expansion plan. In March, Musk announced that SpaceX (already merged with xAI) and Tesla will invest $25 billion in Austin, Texas to build two top-tier chip factories: one dedicated to mass-producing chips for automobiles and humanoid robots, and the other focused on R&D and manufacturing for space AI data centers.
Musk personally characterized Terafab as the “most epic chip manufacturing initiative in history,” integrating logic, memory, advanced packaging, lithography, manufacturing, and testing within a single park-based ecosystem, with the goal of producing up to 1 terawatt (terawatt) of compute power each year.
Intel confirmed the partnership on the X platform, claiming that the company’s technical capabilities will accelerate the realization of Terafab’s goals. CEO Lip-Bu Tan further endorsed it personally after the handshake photo was released: “Musk has impeccable track record in redefining the industry—Terafab represents a step-change transformation in how silicon logic, memory, and packaging technologies are manufactured.”
Intel is proud to join the Terafab project with @SpaceX, @xAI, and @Tesla to help refactor silicon fab technology.
Our ability to design, fabricate, and package ultra-high-performance chips at scale will help accelerate Terafab’s aim to produce 1 TW/year of compute to power… pic.twitter.com/2vUmXn0YhH
— Intel (@intel) April 7, 2026
For Intel, this deal is especially crucial. Intel Foundry (the foundry business) racked up cumulative losses of up to $10.3 billion in 2025. It has long been pressured by TSMC, and after its AI chip cooperation plans hit obstacles, SoftBank shifted its focus back to TSMC, casting a shadow over Intel’s foundry transformation.
At the same time, during its restructuring, Intel secured investments worth billions of dollars from NVIDIA and the U.S. government. The U.S. government has even leapfrogged to become Intel’s largest shareholder. The Trump administration’s “Made in America” strategy has made Intel an indispensable piece on the geopolitical chessboard. With Terafab’s entry, Intel Foundry finally gets a major customer on a scale that matches its capacity.
Musk’s AI compute map: from xAI’s Colossus cluster to edge inference for Tesla robot Optimus—every link requires massive amounts of custom chips.
If Terafab can land on schedule, Intel could be positioned to enter a niche focused on “U.S. domestic manufacturing, vertical integration, and Musk’s ecosystem.”
Of course, the challenges are just as big. Intel’s process technology has long lagged TSMC by one to two generations; whether it can meet Terafab’s stringent requirements for yield and specifications remains to be verified. But with the dual boost of the Trump administration pushing supply chain reshoring and Musk holding government procurement resources, this may indeed be a good opportunity for Intel to make a comeback.