Intel has signed on to Elon Musk’s Terafab AI chip project, a move that links the once-dominant chipmaker to a sprawling plan meant to feed Tesla’s humanoid robots, cars, and data centers. The market liked it too: Intel shares rose more than 2% after the announcement, a neat reminder that even a turnaround story can get a boost from a famous handshake.
The Intel Terafab chip push gives the company a chance to prove it can do more than survive. It also gives Musk another supplier for an industrial fantasy that now spans ground vehicles, robots, and, apparently, ”AI data centres in space.” Ambitious? Yes. Easy? Not even slightly.
What Intel is bringing to Terafab
Intel said its capabilities will help Terafab reach a target of producing 1 terawatt per year of compute for future AI and robotics work. That is the sort of number that sounds like it was chosen by an engine room, not a spreadsheet. Intel chief executive Lip-Bu Tan praised Musk’s track record and framed Terafab as a new model for how silicon logic, memory, and packaging will be built.
For Intel, that rhetoric matters because the company has spent the last stretch trying to convince investors it can still matter in the AI era. Nvidia has already poured billions into Intel, and the U.S. government is now its biggest shareholder, so every high-profile customer win becomes part of a much larger repair job.
Musk’s Terafab chip plan is bigger than Tesla alone
The Terafab project follows Musk’s earlier remarks that Tesla and SpaceX, now merged with xAI, would build two advanced chip factories in Austin, Texas. One plant is supposed to power cars and humanoid robots. The other is aimed at AI data centres in space, which is either visionary industrial planning or the kind of sentence that makes traditional chip executives reach for aspirin.
Musk had already floated the idea that Tesla could work with Intel, so the deal is less a surprise than the next rung on a ladder he has been describing for months. The difference now is that Intel is no longer just being mentioned as an option. It is in the project.
Why this helps Intel’s turnaround
Tan has been pushing an aggressive restructuring at Intel, including job cuts and asset sales, while trying to revive a foundry business that still loses a lot of money. In 2025, Intel Foundry posted an operating loss of $10.32 billion, even though revenue rose 3%. That gap tells you why a headline like this matters: the company needs real customers, not just better slides.
- Intel shares jumped more than 2% after the announcement.
- Terafab aims to produce 1 terawatt per year of compute for AI and robotics.
- Intel Foundry posted an operating loss of $10.32 billion in 2025.
- The segment’s revenue climbed just 3%.
There is also a timing wrinkle. Intel has said its 18A manufacturing technology may be offered to external customers after being mostly kept for internal use last year. If Terafab turns into a meaningful customer relationship, it gives Intel a showcase at exactly the moment it wants to prove its foundry can attract serious outside business.
The open question is whether this is the start of a broader manufacturing comeback or just a very expensive logo in Musk’s ecosystem. Intel has landed a noisy partner, but the harder part will be converting that into chips, revenue, and, for once, a storyline that doesn’t depend on rescue language.