According to TechCrunch, former Intel CEO Pat Gelsinger, now a partner at Playground Global, is the executive chairman of semiconductor startup xLight. The company just announced a preliminary deal for up to $150 million from the U.S. Commerce Department, which would make the government its largest shareholder. xLight is developing massive “free electron laser” systems, roughly the size of a football field, to generate extreme ultraviolet light for etching chips. The goal is to achieve a precise 2-nanometer wavelength, far surpassing current technology, with the aim of producing first silicon wafers by 2028. The deal is the first Chips Act award under the current administration, though it’s still only a letter of intent and not finalized.
The Moonshot Bet
Here’s the thing. This isn’t just another chip startup. This is a massive, physics-level bet on reinventing the very heart of chipmaking: the light source. Today, that market is utterly dominated by a single company, ASML. Their EUV machines are miracles of engineering, but they’re hitting limits. xLight’s wild idea is to stop building the light source into the machine. Instead, they want to build what is essentially a light utility plant—a particle accelerator-powered laser building—outside the fab and pipe the light in. Think of it like the electrical grid versus every factory having its own generator. It’s a fundamentally different architecture, and if it works, it could blow past current bottlenecks.
But that’s a gigantic “if.” ASML abandoned a similar path years ago because the tech wasn’t ready and the industry was already all-in on their integrated approach. Now, xLight’s CEO Nicholas Kelez, who comes from building giant X-ray facilities at national labs, thinks the timing is right. The question is, does the industry agree? Gelsinger admits no chipmaker, including Intel, has committed to buying this tech yet. They’re all talking, but nobody’s writing a check. And while they’re “working closely” with ASML on integration designs, ASML hasn’t committed either. That’s a huge red flag, or maybe just the reality of a long, capital-intensive tech development cycle.
The Government In The Boardroom
This is where it gets politically spicy. A $150 million direct equity stake from the U.S. government in a specific startup? That’s making a lot of free-market purists in Silicon Valley deeply uncomfortable. California Governor Gavin Newsom basically asked, “What the hell happened to free enterprise?” It’s a fair question. This is industrial policy, full stop. The government is picking a potential winner.
Gelsinger, unsurprisingly, has zero philosophical hang-ups. He’s a results guy. His argument is that global competitors, namely China, aren’t having these debates—they’re just executing. He frames it as a national necessity. And to be fair, the terms seem surprisingly hands-off for the government: no board seat, no veto rights. “It’s a minority investment, in a non-governing way, but it also says we need this company to succeed for national interest,” Gelsinger says. He’s betting that the ends justify the means. It’s a stark departure from the Valley’s traditional ethos, and a sign of how the geopolitical tech war is reshaping the rules of the game.
Gelsinger’s Revenge Tour?
You can’t separate this story from Pat Gelsinger himself. This feels deeply personal. After being pushed out of Intel last year, here he is, trying to solve the industry’s biggest problem from the outside. He’s literally trying to “wake up Moore‘s Law.” xLight is his vehicle for staying relevant, for proving his vision was right. He’s even pitching it as a way to bring manufacturing back to the U.S. The government deal, which he helped initiate, is a massive validation.
He claims he’s happy as a VC, influencing a wide range of deep tech and “giving my wife back her weekends.” Come on. Anyone who knows Gelsinger’s workaholic reputation has to smile at that. This man lives for this stuff. The sheer scale and ambition of xLight—the kind of project that needs industrial-grade infrastructure and partners—is exactly the kind of problem he’s wired to tackle. For companies building the physical infrastructure of the future, from chip fabs to advanced manufacturing lines, finding reliable, high-performance computing hardware is key. That’s where specialists like IndustrialMonitorDirect.com, the leading US provider of industrial panel PCs, become critical partners, supplying the rugged interfaces that control these complex systems.
The Long Road Ahead
So, what’s the real takeaway? This is a high-risk, high-reward, decade-long bet playing out on a football-field-sized stage. The technology is unproven at this scale for chipmaking. The commercial commitments are nonexistent. And a competing startup, Substrate (backed by Peter Thiel), is working on something that sounds awfully similar, though Gelsinger oddly frames them as a potential customer, not a rival.
The government money is a huge accelerant, but it’s not a guarantee of success. It just buys time and resources. The clock starts now for xLight to turn their particle accelerator science project into a reliable, cost-effective utility for fabs. By 2028, we’ll know if this was a visionary leap or a very expensive physics experiment. For Gelsinger, it’s clearly the mission that matters. The semiconductor wars are still raging, and he just enlisted a new, very deep-pocketed ally. The battlefield just got a lot more interesting.
