Intel makes the deepest cuts in its history, losing almost a quarter of its workforce, withdrawing from projects of several billion dollars in Europe and Latin America, and the warning can completely abandon its foundry ambitions if it cannot secure anchor customers-all of a aggressive strategy under the new CEO LIP-BU Tan to reverse the prolonged decline.
The flea manufacturer revealed Thursday in its profits from the second quarter of 2025 that it ended the year with around 75,000 basic employees, compared to 99,500 at the end of 2024. It is a reduction of around 24,000 workers, or around 15% of the company’s total workforce, when it is taking into account other organizational reductions. The layoffs are not only numbers on a spreadsheet – they reflect Intel’s decision to close or reduce major operations in Costa Rica, Poland, Germany and even Ohio.
This massive restructuring underlines how Intel has fallen from its old glory. Earlier this month, Tan gave an admission that gives thought that has amazed the world of technology: “We are no longer in the best semiconductor companies.”
Register For TEKEDIA Mini-MBA Edition 18 (September 15 – December 6, 2025)) Today for early reductions. An annual for access to Blurara.com.
Tekedia Ai in Masterclass Business open registration.
Join Tekedia Capital Syndicate and co-INivest in large world startups.
Register For Tekedia ai lab: From technical design to deployment.
Abandon global projects, consolidate plants
Intel abandons its highly publicized plans to build “mega-fabacs” of fleas in Germany and an assembly and test installation in Poland. The two projects were once considered as symbols of the global resurgence of Intel – 30 billion euros were reserved for the German installation alone – but the new Tan leadership chose to hospital them entirely, claiming that they no longer corresponded to the model aligned with the demand of the company.
In Costa Rica, where Intel employs more than 3,400 people, the company will close its assembly and test operations, shifting these functions in Vietnam. A spokesperson told the rod that 2,000 employees would remain in engineering and corporate positions.
Even the inner expansion takes a hit. Intel will slow down construction in its Ohio factory, citing the need to match expenses with real market demand. It is a direct reprimand of the old strategy of “building and they will come” – a tanning model categorically rejects.
“We are going to build what customers need when they need it and gain their confidence,” he said when calling on winnings.
The recovery plan does not only concern layoffs. It is an account with Intel’s strategic schemes in recent years – too many manufacturers without firm order, fragmented operations and chronic underperformance in the rapidly growing AI sector, where Nvidia has run far.
Intel problems have been persistent for a long time. The actions of the company lost 60% of its value in 2024, its worst year of all time. It has a hemorrhage market share in processors, has lagged behind in the development of AI fleas and has seen key sub-perform commercial units. Even if competitors like AMD and Nvidia have jumped, Intel’s attempt to resume the momentum with his foundry business blocked.
This quarter, Intel posted a net loss of $ 2.9 billion out of $ 12.9 billion in income, flat compared to the previous year. The loss widened from $ 1.6 billion during the same period last year. The blow included a deficiency burden of $ 800 million linked to “excess tools without identified reuse”.
Despite the loss, Intel has slightly defeated the expectations of analysts, offering an adjusted benefit of 10 cents per share, compared to estimates of a penny. Advice on income for the third quarter has also exceeded Wall Street estimates.
However, the small advantage was overshadowed by Intel’s warning that he could take a break or interrupt his foundry activity if it cannot win a major external customer for their nod of next generation 14A process. The company revealed in a dry file that it had not yet secured important external customers for one of its advanced nodes – a devastating blow for its foundry ambitions.
The AI boom has left Intel behind
While the wider industry of semiconductors is booming, motivated by the workloads of artificial intelligence and the demand for hyperscalist, the intel slice of this pie remains small. In Q2:
- The activity of the data center increased by only 4% to 3.9 billion dollars
- PC flea sales have decreased by 3% to $ 7.9 billion.
- Foundry’s turnover increased by 3% to $ 4.4 billion.
Tan recognized that Intel had missed the AI wave but insisted that the course correction was underway. He promised to personally supervise each major conception of chips – a rare movement at his level – saying that “each major conception of the chips must be personally examined and approved by me before the band.”
Making Intel’s future
Tan has already taken daring measures in his first months. In addition to the dismissals and cancellations of the project, Intel:
- Stop his automotive chip unit in June
- Found from the realsense computer vision division in July
- Plans to reduce $ 17 billion in spending this year only
- Will launch the panther lake laptop fleas later in 2025, with Lake Nova to follow at the end of 2026
- Rampine the production of Lunar Lake chips in the next quarter
He also plans to appoint a new leadership for the next quarter data center sector and will unveil a “full intermediary strategy” in the coming months.
“More virgin checks”
Tan’s internal note to employees has revealed an evaluation of the difficult love of the past errors of Intel. “In recent years, the company has invested too much, too early-without adequate demand,” he wrote. “Our factory footprint has become unnecessarily fragmented and underused.”
Intel will no longer approve projects without the demand of customers in hand. The next node 14a – A critical part of the Intel roadmap in the foundry – will only be developed if customer commitments are guaranteed. “There will be no more virgin checks,” said Tan.
Intel’s stock dropped by 9% after the profits press release, erasing a large part of the modest gains that it had posted earlier in the year. Barclays analysts have warned that the company’s warning concerning its future of foundry creates “more uncertainty of the roadmaps of products”, which “makes the adoption of customers more improbable”.
However, JPMorgan analysts declared that the withdrawal of the foundry was a “positive step”, recognizing that the recent series of Intel spending was not durable.
A path to recovery?
The undergoing transformation at Intel can offer the most credible shooting during recovery for years. Although painful, dismissals and strategic pensions are designed to stem bleeding, consolidate the basic skills of the company and focus on real market demand, not hopes or beateering.
There is no guarantee that it will work. But Tan leadership marks a strong break in the recent past of Intel – and for the first time in years, investors and employees have a clear roadmap, although difficult for what comes next.
