Intel shares fell as much as 3% to $115.31 on Monday after Elon Musk confirmed that Taiwan Semiconductor Manufacturing Co. is in discussions to participate in Terafab, his planned semiconductor manufacturing project. The stock pared some of the decline later in the session.

The drop followed a 213% year-to-date rally that had made Terafab expectations an increasingly important component of Intel's foundry turnaround story.

What did Musk actually say?

The catalyst was a weekend post from Musk responding to a report by technology journalist Tim Culpan that TSMC was exploring ways to work with Terafab, including potentially building and operating a facility in Texas.

"Just discussions, but something may come of it," Musk said.

That statement shifts the market narrative even though it does not amount to a deal. Intel had emerged as the first major manufacturer publicly tied to the project. In April, Musk said Tesla planned to use Intel's next-generation 14A manufacturing process for Terafab.

The possibility that TSMC could now take a substantial manufacturing role forces investors to reconsider how much of Terafab's expected economic value belongs to Intel.

Why TSMC changes the Intel story

Musk unveiled Terafab in March as an effort to secure a much larger supply of advanced chips for vehicles, humanoid robots, AI systems and space-based data centers. The project is expected to involve two large fabrication facilities in Texas, plus a smaller research fab at Tesla's Giga Texas campus.

• Target output: one terawatt of computing capacity annually — roughly twice current US capacity
• Estimated capital cost: $5 trillion to $13 trillion, according to Bernstein
• Scope and timing remain uncertain

For Intel, Terafab offered more than revenue. A large, technically demanding customer would validate the manufacturing roadmap at the center of its attempt to rebuild the foundry business against TSMC, the world's dominant contract chipmaker.

The division's economics underline the stakes. Intel Foundry generated $5.8 billion in second-quarter revenue, but outside customers contributed only $293 million, and the unit recorded a $2.1 billion operating loss, according to data cited by Yahoo Finance.

D.A. Davidson analyst Gil Luria told MarketWatch that Intel's stock rise has been supported in part by Terafab expectations. If Intel's involvement ends up limited to providing technology rather than manufacturing volume, investors could reassess the premium attached to the stock.

Does TSMC mean Intel is being replaced?

Not necessarily. Musk has indicated separately that TSMC production would likely come in addition to Intel's supply, according to MarketWatch.

A dual-sourcing structure is common in the semiconductor industry, where large customers reduce dependence on a single manufacturing partner. It would also fit the demand profile of Tesla, SpaceX and xAI, which all need large volumes of processors and advanced chips.

TSMC brings a manufacturing record Intel is still trying to match. Chairman and CEO C.C. Wei said in April that a new fab typically takes two to three years to build, followed by one to two years to ramp production. He added that advanced foundry competition ultimately depends on technology leadership, manufacturing execution and customer trust.

Culpan has reported one possible arrangement: TSMC would own and operate a new factory while Terafab commits capital, production volumes or both. That structure would resemble TSMC's existing joint-venture model in Japan and Germany. These remain reported possibilities, not agreed terms.

A reality check for Intel's rally

The market reaction exposes how much expectation investors have already priced in. With the stock up more than 200% this year, the valuation assumes a successful manufacturing turnaround alongside improvements in core operations.

The TSMC talks complicate that thesis. They do not destroy it. Intel could still secure meaningful Terafab volume in a multi-foundry strategy, and its own process technology and advanced packaging capabilities could give it a role even if TSMC operates part of the footprint.

Execution remains the bigger issue. The 14A process is crucial to Intel's foundry comeback, and Terafab could have served as a high-profile demonstration that the technology is ready for large-scale commercial production. Investors will now wait for evidence that Intel can convert the opportunity into actual manufacturing commitments.