After Lip-Bu Tan became CEO of Intel in March last year, the company’s shares — amid difficulties — were practically at a standstill for seven months, while the chipmaker took a nasty beating in the artificial intelligence race.
But, after strengthening ties with some of the biggest technology giants in the world — and winning over US President Donald Trump — Tan begins his second year in office on a much more positive note. Apple and Tesla show interest in the company’s manufacturing capacity. The processors that Intel produces are back in contention, and the resurgence of optimism that the company will finally benefit from the AI boom has led the stock to renew historic highs.
Before he can deliver what shareholders now expect, Tan needs to promote changes within the 57-year-old company that was once a reference in semiconductor manufacturing. Since taking office, he has spent far more time outside the company than inside and has yet to fully and thoroughly explain internally his plan to fix products and production lines, according to more than a dozen current and former employees, who were not authorized to speak publicly.
The core problems remain, they say: Intel needs products capable of regaining lost market share and factories so good that even rivals are forced to direct billions of dollars in orders to it. Neither is guaranteed.
In his first interview as CEO, Tan said he recognized that the company still has “a long way to go”.
“Intel has the technology, talent and scale to lead again, but leadership is achieved through execution,” he said.
Continues after advertising
Tan admitted that he has initially focused more on customers and said that he is now getting closer to building an internal leadership team that he can truly trust to deliver results. “I want to have a team that I can consider a team,” with a “sense of urgency,” he said. He is targeting the end of June to complete the hiring round. This week, he has already made two new additions.
While building his team, Tan has been working around some of Intel’s biggest problems using his network of contacts. In a crucial White House meeting last August, he turned a public spat with Trump into a deal that made the U.S. government Intel’s third-largest shareholder. To reverse the climate with the president, the executive turned to friends in the sector who could attest to his credibility, including Michael Dell, according to people familiar with the matter.
More recently, Tan entered into a partnership with Elon Musk on a project to build a gigantic factory complex and shake up the chip industry. The agreement was the result of direct conversations between Tan and Musk over time and took much of Intel’s own leadership by surprise, according to these people.
Musk said on Friday that he visited an Intel factory in Oregon this week and posted a photo alongside Tan.
The chipmaker is also in early-stage negotiations to potentially produce the main processors for some Apple devices. THE Wall Street Journal reported on Friday that the two companies reached a preliminary agreement, which caused Intel’s shares to soar again, following a report from Bloomberg Newsat the beginning of the week, about the conversations.
Tan, who has served on more than a dozen semiconductor company boards and personally invested in many others, said he heard from several friends that he shouldn’t take the job at Intel. The Silicon Valley pioneer was experiencing a moment of turmoil, surrounded by doubts even about her survival. The company had let its former manufacturing leadership languish. Its products for PCs and servers, made with outdated production technology, lost market share. And the company missed the biggest opportunity in the history of the semiconductor industry: it was unable to develop an AI accelerator chip capable of competing with Nvidia and was left out of the beginning of the artificial intelligence boom.
Continues after advertising
Former business rivals — including Nvidia CEO Jensen Huang — have come to say that central processing units (CPUs), Intel’s main product, will play a central role in AI data centers. But over the past decade, Intel’s factory capacity has plummeted, to the point where the company has outsourced production of some of its most important chips to TSMC. To prove its reliability, Intel needs to start meeting new technology launch schedules, said Naga Chandrasekaran, who has headed the company’s manufacturing area for almost two years.
Chandrasekaran stated that one of his first goals is to regain “internal clientele” — that is, to get his own product teams back to manufacturing in-house, rather than with third parties. But even that wouldn’t be enough. “Intel’s products alone, even in an extremely successful scenario, are not enough to finance the capital, fill the factories and reach the scale necessary to be truly competitive today in the silicon business,” he said.
He and Tan talk a lot about regaining customers’ trust. “He sits in front of me and tells me what the customer is saying,” said Chandrasekaran, a former Micron executive. “There are ten pages of notes — and there’s no way around it.” Tan has emphasized to potential customers at his factories that they will receive treatment at least equal — if not better — than Intel’s own internal product teams.
Continues after advertising
Tan said he has a plan for where he wants Intel to be in two, five and ten years. “Credibility comes from results,” he said. But in internal communications, although he does not ease up on criticism of the company’s performance, he delegates the details of the plans to executives like Chandrasekaran, according to employees.
Those who know Tan or have worked with him say that his management style mirrors his approach as a venture capital investor. When looking for someone to hire, they don’t want a dossier full of details and spreadsheets, they report. Instead, he prefers a more strategic-level conversation about the state of the industry. If the person makes a good impression, he starts to support him and directs his energy to opening doors for him to succeed, instead of dissecting his strategy or his numbers.
But, in the chip industry, success and failure live precisely in the details.
Continues after advertising
Semiconductor factories that aren’t running near capacity — or even efficiently — can become a huge burden. According to New Street Research, Intel faces costs per chip up to three times higher than those of industry leader TSMC. Most of this difference — more than 40% — is linked to yield, that is, the number of good chips generated in each round of production. Intel’s yield rate is around 65%, compared to more than 80% for the Taiwanese company. Only 8% of the difference in costs is explained by the fact that labor in the US is more expensive.
Even if Intel’s rivals decide to trust Tan, the industry’s unforgiving math makes them hesitant to spend what it would take to migrate, even partially, to a new supplier. They demand proof that the move will pay off immediately — and no one wants to be first, says analyst Louis Miscioscia of Daiwa Capital Markets.
“You want another company to be the one to partner with Intel and go through the hassle,” he said.
Continues after advertising
Within Intel itself, some still advocate that the company be divided, separating manufacturing and product design to accelerate advancement. Tan says this can’t happen anytime soon and there are advantages to keeping both parties connected. In the long term, he says he sees something similar to the model that EMC adopted when operating VMware as a controlled subsidiary.
For now, Tan needs to inject speed into the company. As former CEO Pat Gelsinger summarized: Intel, which once had 99% of the data center processor market, was created to lead, not simply compete.
Kevork Kechichian, who Tan brought in to run the crucial server chip division, said that when he talks to teams behind a deadline in about two weeks, he hears a response he wasn’t used to hearing at Qualcomm, Arm or other companies he’s worked for.
“I asked, ‘What’s the recovery plan?’ And they responded that the recovery plan was… adjust the schedule for another two weeks,” he said.
Getting at least 80% of the organization to believe in the need for urgency — and commit to it — is one of the leadership’s priorities, Kechichian said.
Under the previous management, Intel’s numbers — three straight years of losses and revenue 33% below the 2021 peak — were not presented in such a serious light internally, according to Chandrasekaran.
Investors, who had driven shares higher after capital injections last year, got a reminder in January that the Tan-led “rescue” is still far from complete. The company’s financial performance fell short of projections in part because Intel had not set aside enough capacity to meet the resumption of demand for data center chips and in part because its latest technology, called 18A, is not yet generating as many usable parts as the company would like. Since then, Intel has made adjustments, but it is still not at the same level as the industry leaders.
“I am disappointed that we are unable to fully meet demand in our markets,” Tan said at the time. “My team and I are working tirelessly to increase the efficiency and production of our factories.” “Although yields are in line with our internal plans, they are still below the level that I consider ideal.” Since then, he has said that factories are improving rapidly and that external interest in using Intel’s factory structure is growing.
As much as the semiconductor sector is governed by cutting-edge science and an extremely tough economy, the “momentum” factor also weighs in. For more than 30 years, it was Intel that set the pace. Its annual developer conference in San Francisco was the stage for unveiling the technologies that would drive the rest of the computing industry. An attempt to revive the event under Tan’s predecessor was aborted to cut costs.
Today, that role has passed to Nvidia CEO Jensen Huang, and his GTC conference is what defines the direction of AI computing. The event, however, brought a spark of hope to Intel — not because of a revolutionary technical announcement, but because Huang talked about a future explosion in demand for CPUs.
That’s great for Intel — but only if Tan can act quickly as rivals continue to eat away at its market share.
“As much as I believe in Lip-Bu, I think he has been dealt an extremely difficult hand to play,” said Jon Bathgate, manager at NZS Capital. “But if anyone has a chance to make this work, it’s him.”
© 2026 Bloomberg L.P.
