Intel said it will spend €5 billion, about $5.7 billion, to expand chip manufacturing in Europe, concentrating the investment on its fab complex in Ireland. The same day, TechPowerUp reported that Intel is pulling most production of its Nova Lake processors back from outside contractors, driven by improvements in its own manufacturing yields.
The two announcements, taken together, describe a company trying to make good on a strategy it announced years ago. Intel’s IDM 2.0 plan, unveiled in 2021, promised that the company would both build chips for others and manufacture its own designs at scale again. The strategy was received as a slogan at the time, and Intel’s execution since has been uneven. This week’s moves are the most concrete signs yet that the company is executing on both halves of the promise.
The Irish expansion is the European piece. Intel’s Fab 34 in Leixlip runs some of the company’s most advanced production lines, and the new investment extends the site’s role as Intel’s main European manufacturing hub. The company has also pursued a major fab in Germany, a project that has been delayed repeatedly as Intel has cut costs and reordered its priorities. The €5 billion commitment suggests Ireland, not Germany, has emerged as the anchor of the European strategy.
The Nova Lake shift is the more consequential move. Nova Lake is Intel’s next family of processors for PCs and servers, and previous generations had leaned heavily on TSMC for the most advanced manufacturing. TechPowerUp reported that most Nova Lake production will now stay inside Intel, a decision made possible by yield improvements on Intel’s own nodes. For a company that has spent years outsourcing its most important silicon, bringing Nova Lake home is a statement that its fabs are competitive again.
The economics favor insourcing, if the yields hold. Outsourcing to TSMC costs Intel margin on every chip and hands its biggest rival both revenue and negotiating power. Manufacturing internally means higher fixed costs but better gross margins at scale, and Intel’s turnaround plan depends on exactly that math. The company’s executives have said repeatedly that the road back to industry-standard margins runs through its own fabs, and Nova Lake is the first full test of that claim.
The timing is not accidental. Intel has a new chief executive and a restructuring under way, and the company has staked its recovery on the 18A manufacturing node and the customers it can win with it. Bringing a major product family in-house demonstrates to those customers that Intel’s foundry business can handle its own toughest designs, a form of proof that slides and benchmarks cannot provide.
The scale question remains. €5 billion is a large investment by most standards, but Intel’s competitors spend at a different level: TSMC’s annual capital expenditures have run to roughly ten times that amount, and the gap between the two companies’ spending is the single best explanation for the gap in their manufacturing positions. Analysts said Intel’s European expansion, welcome as it is, leaves the company an order of magnitude short of what a full catch-up would require.
The political context helps explain the size. The European Union’s Chips Act was designed to double Europe’s share of global chip production, and subsidies from Brussels and member states have lowered the cost of Intel’s European projects. The company has been careful to frame its investments as European commitments rather than American exports, and the Irish expansion fits the pattern of a company that has learned to make its capital spending work with government money.
The U.S. government’s own investment frames the company’s choices. Washington has committed billions through the CHIPS Act to rebuild American semiconductor manufacturing, and Intel is the largest single recipient of those funds, with its Arizona and Ohio fabs built partly on federal money. The company’s turnaround has become a national project in a way that no other chip maker’s has, which gives it access to capital but also ties its decisions to political expectations. Every expansion, including the Irish investment, is now read in Washington as well as on Wall Street.
The market’s reaction has been muted, which is itself a signal. Intel shares have been volatile through the turnaround, and investors have learned to wait for results rather than announcements. The Nova Lake decision is a test that will play out over years, in yields, margins, and the foundry customers Intel can win. This week’s news sets the terms of the test; it does not answer it.
For the industry, the significance is broader. If Intel can bring its flagship processors home and make money doing it, the balance of the foundry market shifts, and TSMC’s pricing power faces a real challenge for the first time in a decade. If the plan stumbles, the company’s European plants and its internal manufacturing will be remembered as the last big bet of a fading giant. The outcome will be decided by yields, and the first data point is already in: Intel believes it can make Nova Lake itself.
