Intel Corporation has spent decades as the archetype of American semiconductor dominance. It designed chips. It built them. It sold them. That vertically integrated model made the company a titan. Then it didn’t.
Now, under a restructured leadership team still executing the strategic vision that former CEO Pat Gelsinger set in motion, Intel is attempting something that no Western chipmaker has pulled off in the modern era: building a world-class contract chip manufacturing business from scratch while simultaneously trying to keep its own product lines competitive. The potential upside, according to a recent analysis by Yahoo Finance, could be as large as $500 billion in enterprise value — if Intel Foundry Services can capture even a modest share of the booming global demand for advanced semiconductor fabrication.
That’s a staggering number. It’s also deeply conditional.
The Architecture of a $500 Billion Thesis
The bull case for Intel Foundry rests on a convergence of structural forces that, individually, are each powerful. Together, they form what proponents argue is a once-in-a-generation opportunity for the company to reinvent itself as the Western world’s answer to Taiwan Semiconductor Manufacturing Company.
First, there’s the geopolitical dimension. The CHIPS and Science Act, signed into law in August 2022, allocated $52.7 billion in subsidies and incentives to onshore semiconductor manufacturing in the United States. Intel has been the single largest beneficiary. The company has secured up to $8.5 billion in direct funding from the Commerce Department, plus billions more in loan guarantees and tax credits. This isn’t charity — it’s a national security calculation. With TSMC fabricating the vast majority of the world’s most advanced chips on an island 100 miles from mainland China, the strategic vulnerability is obvious. Washington wants alternatives. Intel is the most credible domestic candidate.
Second, the demand picture is extraordinary. Artificial intelligence workloads are driving an unprecedented surge in demand for advanced semiconductors. Nvidia can’t get enough supply from TSMC. Neither can AMD, Apple, Qualcomm, or the growing roster of hyperscalers designing their own custom silicon. The total addressable market for foundry services is projected to exceed $200 billion annually by the end of the decade, up from roughly $120 billion today, according to estimates from industry research firms including TrendForce and Gartner.
Third — and this is where the $500 billion figure comes from — if Intel Foundry can capture even 10% to 15% of that market at scale, the revenue and margin profile would transform the entire company. Apply TSMC-like multiples to a business generating $20 billion to $30 billion in annual foundry revenue at 40%-plus gross margins, and you get a standalone valuation that dwarfs Intel’s current market capitalization, which sits around $110 billion as of mid-2025.
But the gap between the thesis and the execution is enormous. And Intel knows it.
Intel Foundry Services lost $7 billion on an operating basis in 2024. That’s not a typo. The unit is hemorrhaging cash as Intel simultaneously builds out new fabs in Ohio, Arizona, and Germany while trying to bring its advanced process nodes — Intel 18A and beyond — to manufacturing readiness. The company has acknowledged that IFS won’t reach operating breakeven until 2027 at the earliest, and profitability on a sustained basis likely won’t arrive until the end of the decade.
The spending is breathtaking. Intel’s capital expenditure plans call for roughly $25 billion to $28 billion annually over the next several years, a figure that makes it one of the largest infrastructure investors in the world, period. The Ohio fab complex alone, which Intel has branded as a mega-site in New Albany, represents a $28 billion commitment when fully built out. The Magdeburg, Germany facility — Intel’s first major European fab investment — carries a price tag north of €30 billion, though a significant portion of that is backstopped by German and EU subsidies.
These are bets that take years to pay off. Semiconductor fabs don’t generate revenue overnight. They take three to four years to construct, another year or more to qualify, and several more years to ramp to full production volumes. Intel is essentially asking investors to fund a decade-long transformation with no guarantee that customers will show up in sufficient numbers.
So far, the customer pipeline is encouraging but thin. Intel has announced foundry agreements with Microsoft and the U.S. Department of Defense, among others. Microsoft’s commitment is particularly significant — the software giant is designing custom AI accelerators and needs a non-TSMC fabrication partner for strategic diversification. But the details of these deals remain opaque. Volume commitments, pricing terms, and technology node specifics have not been publicly disclosed in any meaningful detail.
The Technology Gap That Must Close
None of the financial projections matter if Intel can’t deliver competitive manufacturing technology. This is the crux of the entire strategy.
TSMC’s dominance isn’t an accident. The Taiwanese foundry has spent 30 years refining its manufacturing processes, building customer relationships, and creating a virtuous cycle where scale begets yield improvements, which beget more customers, which beget more scale. Intel is trying to compress that timeline dramatically.
The company’s roadmap is ambitious. Intel 18A, the process node that Intel is positioning as its foundry flagship, uses gate-all-around transistor architecture and backside power delivery — two innovations that Intel argues will give it a technical edge over TSMC’s competing N2 node. If 18A performs as advertised, it would represent the first time in years that Intel has been at or near the leading edge of manufacturing technology.
Early indications are mixed. Intel has demonstrated working silicon on 18A, and the company has said that yield improvements are tracking ahead of internal targets. But “ahead of internal targets” is a relative statement from a company that spent the better part of five years missing its own technology deadlines. The credibility deficit is real. Intel was supposed to be manufacturing at 10nm in volume by 2016. It didn’t get there until 2019. The 7nm debacle that followed ultimately cost former CEO Bob Swan his job and brought Gelsinger back to the company.
Gelsinger’s departure in late 2024 added another layer of uncertainty. He was the architect of the IDM 2.0 strategy — the plan to transform Intel into both an internal chipmaker and an external foundry. His exit raised questions about whether the board and the new leadership team would maintain the same level of commitment to the foundry buildout, particularly given the massive losses.
The answer, so far, appears to be yes. Co-CEOs David Zinsner and Michelle Johnston Holthaus have reaffirmed the foundry strategy, and the company hasn’t pulled back on its major capital projects. But the pressure is mounting. Intel’s stock has significantly underperformed the broader semiconductor index over the past three years, and activist investors have been circling.
There’s also the competitive reality to contend with. TSMC isn’t standing still. The company is building its own U.S. fabs in Arizona — the first of which is already producing chips — and is investing heavily in advanced packaging technologies like CoWoS that are critical for AI chip manufacturing. Samsung, Intel’s other major competitor in advanced foundry, has struggled with its own yield issues at 3nm but remains a formidable player with deep pockets and Korean government backing.
And then there’s the customer acquisition challenge. Chip designers are notoriously conservative about switching foundries. Moving a design from TSMC to Intel isn’t like switching cloud providers — it requires years of co-development, extensive qualification testing, and a willingness to bet hundreds of millions of dollars in development costs on a manufacturing partner’s ability to deliver. Most fabless chip companies would rather pay a premium to stay with TSMC than take the risk of qualifying a new foundry, even one backed by the U.S. government.
This is why Microsoft’s involvement matters so much. If a customer of that scale and sophistication is willing to commit to Intel Foundry, it sends a powerful signal to the rest of the industry. But one marquee customer does not a foundry business make.
The Valuation Paradox
Here’s the uncomfortable math. Intel’s current market capitalization implies that the market is assigning little to no value to the foundry business. In fact, given the scale of the losses and capital commitments, it’s possible that the market is treating IFS as a net negative — a drag on the company’s more established product divisions.
That creates an asymmetric setup for investors willing to take a long view. If Intel Foundry fails to gain traction, the downside is significant but arguably already priced in. If it succeeds — if 18A works, if customers come, if the government subsidies keep flowing, if the geopolitical environment continues to favor onshoring — the upside is genuinely transformational.
The $500 billion figure cited by Yahoo Finance represents the upper bound of that optimistic scenario. It assumes Intel Foundry achieves something approaching TSMC’s margin profile and commands a comparable valuation multiple. That’s a heroic assumption. TSMC’s margins are the product of decades of operational excellence and a near-monopoly on the most advanced nodes. Intel is years away from anything resembling that position.
A more moderate bull case might value a successful Intel Foundry at $100 billion to $200 billion — still a massive number relative to Intel’s current enterprise value, and enough to justify the investment thesis for patient capital. But even that scenario requires flawless execution over a multi-year timeline, something Intel has not demonstrated in recent memory.
The bears have a simpler argument. Intel has a long history of overpromising and underdelivering on manufacturing technology. The foundry business model requires a fundamentally different culture than Intel’s traditional integrated approach — one focused on serving external customers rather than internal product teams. Intel’s organizational DNA may not support that transition. And the capital requirements are so enormous that any stumble could force the company to raise dilutive equity or take on unsustainable debt.
Both sides have a point. And that tension is exactly what makes Intel one of the most consequential bets in the semiconductor industry right now.
The next 18 months will be decisive. Intel 18A needs to reach high-volume manufacturing readiness. The Ohio and Arizona fabs need to hit their construction milestones. And the company needs to announce additional foundry customers with real volume commitments — not just memoranda of understanding or vague partnership announcements.
If those things happen, the $500 billion thesis starts to look less like fantasy and more like a plausible outcome. If they don’t, Intel will face increasingly difficult questions about whether the foundry strategy was the right bet — or just the most expensive one.
Either way, the stakes extend far beyond one company’s balance sheet. Intel Foundry is, in many ways, a proxy for whether the United States can rebuild a world-class semiconductor manufacturing base on its own soil. The answer to that question will shape technology supply chains, national security calculations, and the global balance of economic power for decades to come.
No pressure.
WebProNews is an iEntry Publication