Why INTC Stock’s Terafab Partnership With Elon Musk Changes the Foundry Story Completely

Elon Musk and Intel CEO Lip-Bu Tan were seen shaking hands in the photo that went viral on Tuesday. It was posted to X in the casual manner that big announcements are now made, not through an earnings call or press release but rather through a social media post that causes a stock price to rise by 4% in a matter of hours. Intel announced its participation in Musk’s Terafab initiative, which aims to construct a vertically integrated chip manufacturing facility in Austin, Texas, that would eventually provide semiconductors for SpaceX, Tesla, and xAI. INTC’s stock ended the day at $52.91, up 4.19%, not far from its 52-week high of $54.60. The chart alone reveals a story worth closely examining for a company that was trading below $18 at its lowest point in the previous year.

The Terafab project is so ambitious that some analysts have referred to it as a “herculean task”; Morgan Stanley used this exact term when evaluating the project, predicting that any significant chip output wouldn’t be available until at least the middle of 2028. In order to supply semiconductors to Tesla’s autonomous robotaxi program, the Optimus humanoid robot platform, and SpaceX’s growing network of AI-enabled satellites, Musk intends to produce about one terawatt of computing capacity annually.

Intel Corporation

intc.com ↗

There is a long history of ambitious compute projects falling years behind their projected timelines in the automotive, semiconductor, and aerospace industries, so it is truly uncertain if that timeline will hold. However, the announcement of the partnership is more significant to Intel as a signal than as a near-term revenue event: at a time when Intel’s foundry business sorely needed that kind of high-profile endorsement, one of the world’s most closely watched technology ecosystems selected Intel as a fabrication and packaging partner.

It’s important to recall Intel’s position eighteen months ago. Due to the company’s near-complete failure to capitalize on the surge in demand for AI chips, Nvidia was able to acquire data center silicon at a scale that Intel, with its previously dominant market position, lacked the credibility to swiftly compete for. The stock dropped to a 52-week low below $18, which would have seemed unimaginable in the days when Intel’s name was synonymous with the interior of every significant computer on the planet. After an aggressive but challenging tenure attempting to revitalize the company’s manufacturing capabilities, CEO Pat Gelsinger left in late 2024.

Following Lip-Bu Tan’s takeover in March 2025, there has been a more subdued and methodical change in strategy, including cost reductions, targeted investments, and a conscious attempt to reposition Intel’s foundry services as a US-based alternative to TSMC for clients who either want supply chain diversification or are under regulatory pressure to source chips domestically.

One of the more peculiar aspects of contemporary industrial policy is the US government’s role in that narrative. As of late March, the federal government owned about 8.4% of Intel’s outstanding shares after the Trump administration acquired an equity stake in the company worth about $9 billion.

The justification was simple: Intel is the only American business with the infrastructure and technical know-how to challenge TSMC’s hegemony at advanced nodes, and domestic semiconductor manufacturing capacity is a national security concern. The market is still figuring out whether the government’s support speeds up customer wins or complicates them with regulations. Practically speaking, it has given Intel a financial buffer that has enabled Tan to make choices that would have been challenging to defend in a more precarious situation, such as the $14.2 billion repurchase of the Irish semiconductor facility that it had previously sold to Apollo Global Management in a more dire situation.

The Irish buyback is a signal in and of itself. The decision to repurchase a facility at a premium after selling it during a period of financial hardship speaks volumes about management’s current confidence in the company’s future. The picture that emerges around Intel in early 2026 is one of a company building partnerships and assets rather than selling them, as evidenced by the Terafab partnership, reports of ongoing talks with Google and Amazon for advanced packaging services, and a planned investment increase in AI chip startup SambaNova Systems to roughly 9%. Although revenue was still down 4.1% year over year and the operating margin was still negative, hovering around minus 4.25%, the Q4 2025 earnings showed an EPS beat—$0.15 actual versus $0.08 estimated.

Seeing the INTC chart rise from $17 to $52 over the course of about a year gives me the impression that the market is pricing in a story that hasn’t quite come to pass but is getting harder to ignore. Twenty-six analysts rate the stock as Hold, six as Sell, and only five as Buy. This is an exceptionally conservative consensus for a stock that has almost tripled from its lows and is gaining foundry clients at a rate that would have seemed unlikely eighteen months ago.

The fact that the average price target of $46.19 is actually lower than the current stock price indicates that either the stock is outperforming the fundamentals or the analyst consensus has not kept up with the rapid change in the narrative. The market will be watching closely for any details regarding Terafab order depth, packaging capacity utilization, and the timeline for turning discussions with Google and Amazon into signed agreements. The next concrete data point will be earnings on April 23. The question is whether the catalysts are building up quickly enough to propel the stock through until then, as it is currently slightly below its 52-week high.