A year after being fired from Intel, Pat Gelsinger is still waking up at 4 a.m., still in the thick of the semiconductor war — just on a different battlefield. Now a general partner at venture firm Playground Global, he is working with 10 startups. But one portfolio company has captured a large portion of his attention: xlightIn a semiconductor startup that announced last Monday that it had secured an initial deal of up to $150 million from the US Commerce Department, the government is set to become a meaningful shareholder.
It’s a nice feather in the cap for Gelsinger, who spent 35 years at Intel over two terms before the board showed him the door late last year. Lack of confidence In his change plan. But the xLight deal also shines a light on a trend that’s making people in Silicon Valley quietly uneasy: the Trump administration taking equity stakes in strategically important companies.
“What happened to free enterprise?” California Governor Gavin Newsom ask At a speaking event this week, capturing the malaise sweeping through an industry that has long prided itself on its free-market ethos.
Speaking at one of TechCrunch’s StrictlyVC events at Playground Global, Gelsinger — who is xLight’s executive chairman — wasn’t bothered by the philosophical debate. He’s more focused on his bet that xLight can solve what he sees as the semiconductor industry’s biggest hurdle: lithography, the process of etching microscopic patterns onto silicon wafers. The startup is building massive “free electron lasers” powered by particle accelerators that could revolutionize chip manufacturing. If the technology works at scale, that is.
Pat Gelsinger
You know, I have this long-term mission to continue to see Moore’s Law in the semiconductor industry,” Gelsinger said, referring to the decades-old principle that computing power should double every two years. “We think this is the technology that will trigger Moore’s Law.”
The Xlite deal is the first Chips and Science Act award in Trump’s second term, using funds earmarked for early-stage companies with promising technologies. Notably, the deal is currently in the objective stage, meaning it has not been finalized and details may still change. When pressed on whether the funding could double the amount announced — or perhaps never materialize at all — Gelsinger was coy.
“We have agreed in principle on the terms, but as with any of these agreements, there is still work to be done,” he said.
TechCrunch event
San Francisco
|
October 13-15, 2026
The technology Xlight is pursuing is quite serious in both scale and ambition. The company plans to build machines about 100 meters by 50 meters – the size of a football field – that will sit outside a semiconductor fabrication plant. These free electron lasers will produce extreme ultraviolet light at wavelengths as precise as 2 nanometers, much more powerful than the 13.5 nanometer wavelength currently used by ASML, the Dutch giant that completely dominates the EUV lithography market.
“About half the capital goes into lithography,” Gelsinger explained of the entire semiconductor industry. “In the middle of a lithography machine is light… [and] This ability to continue to innovate for shorter wavelength, high energy light is the essence of being able to continue to innovate for more advanced semiconductors.
Leading the way is Nicolas Kelleys, whose background is unusual for the semiconductor world. Prior to founding xLight, Kelage led quantum computer development efforts at PsiQuantum (a Playground Global portfolio company) and spent two decades building large-scale X-ray science facilities at national labs including SLAC and Lawrence Berkeley, where he was the lead engineer for the Linac Coherent Light Source.
So why is it effective now when ASML abandoned the same approach almost a decade ago? “The difference was the technology wasn’t as mature,” explained Kelage, who was speaking at the event with Gelsinger. At the time, only a few extreme ultraviolet lithography (EUV) machines existed, and the industry had already sunk billions of dollars into the current technology. “It was not time to take on something completely new and orthogonal.”
Now, with EUV ubiquitous in lead-edge semiconductor manufacturing and existing light source technology hitting its limits, the times look better. The key innovation, according to Kelage, is to use light as a utility instead of building it into every machine. “We get away from building an integrated light source with the tool, which is what it is [ASML does] Now and that fundamentally prevents you from making it smaller and less powerful,” he said. We build out of the fab at utility scale and then distribute.”
The company aims to produce its first silicon wafer by 2028 and have its first commercial system online by 2029.
Naturally, there are hurdles, though right now, competing directly with ASML doesn’t seem to be one of them. “We’re working very closely with them to basically design how to integrate with an ASML scanner,” Kelage said. “So we’re working with both of them and their providers, [like] Jeez, who does their optics.”
Asked if Intel or other major chipmakers have committed to buying Xlight’s technology, Gelsinger said they haven’t. “No one has committed yet, but everyone on the list you’d expect is in the works and we’re in deep conversations with all of them.”
Meanwhile, the competitive landscape is heating up. In October, Substrate — a semiconductor manufacturing startup backed by Peter Thiel — announced it had raised $100 million US chip fabs to develop, including an EUV tool that sounds similar to xLight’s approach. Gelsinger, however, does not see them as direct competition. “If Substrate is successful, they could be a customer for us,” he said, adding that Substrate is focused on building a full-stack lithography scanner that will eventually require a free electron laser, which xLight is developing.
Gelsinger’s relationship with the Trump administration adds another layer to the story. He brought xLight to Commerce Secretary Howard Lutnick in February before Playground funded the startup and Lutnick was confirmed. At that point, Kelage said, he had already spent more than a year pitching xLight to the government as a way to bring chip manufacturing back to the U.S., but the new system drew criticism from some who viewed the administration’s approach as excessive.
Gelsinger framed it as unnecessary, necessary for national competition. “I measure it by results,” he said. “Does this drive the outcomes we want and do we need to revive industrial policies? Many of our competitive countries don’t have this kind of debate. They are moving forward with the policies they need to achieve their competitive outcomes.”
He points to energy policy as another example. “How many nuclear reactors are being built in the US today? Zero. How many are being built in China today? 39. Energy policy in a digital AI economy equals the country’s economic power.”
For xLight, government participation has minimal strings attached. The Commerce Department will not have veto rights or board seats, Kelage said (pictured above). “There is no right to information, nothing,” Gelsinger added. It’s a minority investment, in a non-governing way, but it also says that we need this company to succeed in our national interest.
xLight has raised $40 million from investors including Playground Global and plans to raise another fund next month, in January. Unlike a fusion or quantum computing startup that needs billions, Kelage says xLight’s path is more manageable. “It’s not fusion or quantum,” he said. “We don’t need billions.”
The company has signed a letter of intent with New York to build its first machine at the New York Create site near Albany, though that deal also needs to be finalized.
For Gelsinger, xLight is clearly more than just another portfolio company. It’s an opportunity to cement his relevance in the semiconductor industry he helped create, even if his methods put him at odds with Silicon Valley’s traditional ethos.
When asked about navigating his policies in the current political environment, Gelsinger retreated to a more technocratic view of corporate leadership—where money comes from the US government, administrations are temporary, and CEOs must stay above the fray.
CEOs and companies should not be Republicans or Democrats,” he said. “Your job is to meet the business objectives, to serve your investors, to serve your shareholders. That’s your purpose. And as a result, you have to be able to figure out which policies are beneficial on the R side or which policies are beneficial on the D side and navigate through those.”
“Taxpayers will do well,” he added separately, of that $150 million from the Trump administration.
When asked if working across 10 startups was enough for someone who used to run Intel, Gelsinger insisted. “Absolutely. The idea that I can now make an impact across such a wide range of technologies — I’m a deep tech guy who I am. My mind is stretched out here, and I’m just grateful that the Playground team let me join them and let me make them smarter and be a rookie venture capitalist.”
He paused, then added with a laugh: “And I gave my wife back her weekend.”
It’s a nice thought, although anyone who knows Gelsinger’s reputation as a workaholic may wonder how long this arrangement will last.
Pat Gelsinger wants to save Moore’s Law, with a little help from the Feds