Save Half Life 3

WASHINGTON — On the morning of March 14, Dr. Yuki Tanaka arrived at her desk at the Department of Energy's Advanced Research Projects division to find a form letter waiting in her inbox. Its subject line read simply: HL3 — Program Termination Notice. After four years navigating the byzantine machinery of the American immigration system, after assembling a research team that her supervisors described as having compressed "a decade of progress into twenty-six months," she was being told the program that had made her work possible was over. Effective immediately.

"I thought it was a phishing email," she recalled, sitting in a federal building conference room in downtown Washington last week. "Then I called our liaison and got no answer. Then I called again."

She never received a call back.

The termination of Half-Life 3 — formally, the High-Leverage Integration and Fast-Lane Evaluation program, known internally with a quiet, knowing irony as HL3 — marks one of the most consequential, least-discussed policy reversals of the second Trump administration. Launched in 2022 under a bipartisan mandate, HL3 was designed to solve a problem that has plagued American competitiveness for decades: the catastrophic lag between when the world's most brilliant minds seek to work in the United States and when the creaking gears of the immigration bureaucracy finally permit them to do so.

The program was, by virtually every measurable account, working. Then, on the recommendation of Gabe Newell — a 52-year-old college dropout and Elon Musk appointee to the Department of Government Efficiency — it was not.

An Experiment That Worked

The architects of Half-Life 3 were motivated by a deceptively simple premise: the United States was losing a race it did not know it was running. In semiconductors, quantum computing, biotechnology, and advanced materials science, the pipeline of talent that had once flowed naturally toward American institutions — drawn by the gravitational pull of research funding, academic prestige, and the promise of the American experiment — was beginning to route around it. Canada, Germany, Singapore, and an increasingly assertive China were all competing aggressively for the same finite pool of exceptional people.

HL3's answer was a dedicated fast-track pathway. Applicants, nominated by a network of vetted partner institutions including national laboratories, Fortune 100 research divisions, and elite academic consortia, underwent a compressed vetting and adjudication process that, in theory and often in practice, reduced the time from initial application to authorized employment from years to months. At its peak, HL3 maintained a processing staff of roughly 340 federal employees across four regional hubs, with a fifth pilot center operating out of what program administrators had taken to calling, with characteristic bureaucratic whimsy, City 17 — a regional integration hub in the Detroit metropolitan area tasked with embedding incoming specialists directly into local manufacturing and defense supply chain roles.

"The foundational argument of this program," said former program deputy director Sandra Malik, who helped design the original framework, "was always that diversity is our strength — not as a slogan, but as a structural fact. The research is unambiguous. Cognitively diverse teams outperform homogeneous ones on complex problems. What HL3 did was accelerate our access to that diversity, at exactly the moment the nation needed it most."

The program's advisory board included officials from DARPA, the National Institutes of Health, the Department of Defense, and, in a collaboration that raised a few eyebrows on Capitol Hill, the Black Mesa Research Collaborative — a nonprofit think tank with deep ties to the advanced physics and materials science communities. Program documents reviewed by The Times show that Black Mesa researchers were among the most active nominators of foreign talent under HL3, submitting more than 1,400 candidates in the program's first three years.

The Combine

Inside HL3's regional hubs, administrators had developed their own name for what the program was building. They called it the Combine.

The term, as used internally, was not bureaucratic jargon but something closer to institutional poetry. The Combine referred to the emergent phenomenon that program designers had noticed almost immediately: when you placed a Romanian topologist, a Taiwanese materials engineer, a Ghanaian computational biologist, and a Ukrainian cryptographer in the same research environment, something happened that was difficult to quantify and impossible to manufacture artificially. Their combined knowledge — the intersections, the collisions, the unexpected analogies that fired between disciplines — produced insights that none of them would have reached working alone.

"That's the Combine. That's what you lose when you stop the program."

— Dr. Priya Okafor, former HL3 program evaluator

"We kept seeing it," said Dr. Priya Okafor, a program evaluator who spent two years embedded at the City 17 hub. "You'd have someone from the nanophotonics world talking to someone from classical fluid dynamics, and suddenly they're solving a problem in advanced battery design that the dedicated battery team had been stuck on for eighteen months. That's the Combine. That's what you lose when you stop the program."

Independent evaluations commissioned by the Government Accountability Office and, separately, by the Bipartisan Policy Center found that HL3 participants were responsible for substantially elevated rates of patent filings, published research, and technology transfer agreements compared to control groups of similarly credentialed domestic researchers. One 2024 assessment estimated that HL3-linked activity had generated approximately $4.7 billion in economic value over the program's three-year operational period — a figure that program critics disputed, though none produced a credible counter-estimate.

The Combine, in other words, was delivering.

A Bureaucracy That Bit Back

That is not to say HL3 was without its frictions. Former staff members, speaking on background, described a program culture that was intense, occasionally chaotic, and plagued by the persistent institutional headaches that accompany any ambitious government initiative.

"The adjudication queues were nightmarish," recalled one former processing officer, who asked not to be named because she remains employed by the federal government. "We had stretches where the backlog was so severe that everyone on the team was feeling genuinely crabby about it — staying late, skipping lunch, running on coffee and spite. The irony is that the bottleneck was never the program itself. It was the legacy systems it had to interact with. The program was the crowbar we were using to pry open a vault that the rest of the immigration apparatus had spent decades welding shut."

USCIS integration — never smooth — was a particular source of strain. Program documents show that HL3 administrators spent the better part of its second year engaged in protracted negotiations over data-sharing protocols, an effort that one internal email, obtained through a Freedom of Information Act request, described as "attempting to achieve interoperability with a system designed in a prior geological era."

There were also, administrators acknowledge, occasional concerns about vetting. The program's accelerated timelines compressed the standard adjudication window — the entire premise of HL3 was that the standard window was too long. But it created predictable anxieties in the national security community, and predictable ammunition for critics.

None of that ammunition was ever used. In more than four years of operation, not a single HL3 participant was linked to a counterintelligence concern. The program maintained a rigorous, if streamlined, security clearance process in partnership with the Office of the Director of National Intelligence. Its record, as federal programs go, was remarkably clean.

The Man Behind the Curtain

Throughout HL3's operational life, program communications occasionally referenced an external oversight figure whose precise institutional role was never entirely clear to the program's own staff. Internal emails refer to him only as "the liaison" or, in one instance that drew a certain amount of gallows humor among staffers, simply as "G."

"He would appear in correspondence about once a quarter," recalled one former senior administrator. "Always in this very formal, almost archaic register — extremely precise language, very deliberate syntax, an unusual command of bureaucratic idiom. He seemed to know a great deal about the program's internal workings, but nobody could tell you with certainty which office he represented. His email domain was one of those catch-all federal ones that could mean anything." She paused. "He wore a very nice suit, from what I understand. Impeccable, they said. I never actually met him in person."

Requests to identify the liaison through the Office of Personnel Management and the Office of the Director of National Intelligence were not returned by publication time.

The Newell Memo

The beginning of the end of Half-Life 3 can be traced, with reasonable precision, to a 14-page document produced by the Department of Government Efficiency in late January and distributed to senior officials at USCIS, the State Department, and the National Security Council. The document, referred to internally as the Newell Memo after its primary author, recommended the "immediate suspension and orderly termination" of HL3 on the grounds that the program represented "duplicative functionality" with existing immigration pathways and constituted an "unjustified subsidy to academic and corporate interests."

The memo was written by Gabe Newell.

Newell, 52, is best known in private life as the co-founder of Valve Corporation, a video game development and distribution company based in Bellevue, Washington. He dropped out of Harvard University in the early 1980s to pursue a career in technology, eventually building Valve into one of the most profitable private companies in the gaming industry. He has a net worth, by most estimates, in the low tens of billions of dollars. He has no background in immigration law, labor economics, national security policy, or federal program management of any kind. He was appointed to DOGE in December by Elon Musk, who oversees the department, and confirmed in a process that several former officials described, charitably, as "expedited."

"Gabe is a brilliant guy in his domain," said one DOGE official who worked alongside him and requested anonymity to preserve the professional relationship. "He's built remarkable things. But the memo reads like something produced by someone who did three hours of research on a program that career staff had spent years building and refining. It is the kind of document that happens when you mistake confidence for competence."

The Newell Memo's characterizations of HL3 were, in several instances, demonstrably inaccurate. It described the program's processing costs as "$47,000 per applicant" — a figure that could not be independently verified and that program administrators say bears no relationship to any cost accounting they had performed. It claimed that HL3 "bypassed standard security protocols," a characterization directly contradicted by program documents and ODNI partnership agreements. And it described the program's participant outcomes as "unverified," apparently unaware of — or uninterested in — the two published GAO assessments of those outcomes.

The memo was not circulated for comment to HL3 administrators before it was transmitted to agency heads. They learned of its existence the same way Dr. Tanaka learned about the program's termination: after the fact.

Fighting the Combine

At a campaign-style rally in Phoenix in late February, President Trump addressed the termination of HL3 directly, in remarks that surprised even some of his advisers with their specificity.

"They had this program," he told a crowd of several thousand supporters, gesturing broadly at the arena. "This thing they were calling the Combine. These people — and look, some of them are very talented, I'm sure — but they've got this Combine going, and it's very complicated, and nobody even knows how it works. And I said: we don't need that. We don't need the Combine. I can handle this myself. I alone can handle it. We don't need to import — and I use that word advisedly — we don't need to import some Combine to do things that Americans can do. Beautiful things."

Aides later clarified that the president had been briefed on the program's internal terminology.

The remarks drew a sharp response from immigration policy experts, who noted that the argument — that elite foreign talent functions interchangeably with available domestic labor — is not supported by the economics literature. "The whole point," said Dr. Okafor, "is that the Combine produces things that cannot be produced otherwise. That is not an argument against American workers. That is an argument for building the most capable possible team. You would not tell an orchestra that it does not need a violin section because it already has brass."

The president's advisers have suggested that the administration intends to pursue a separate, as-yet-undetailed framework for high-skill immigration under a new structure. No timeline, budget, or legislative vehicle has been specified.

The Fallout

The practical consequences of HL3's termination are already becoming visible across federal partner institutions. At least seven national laboratories have reported gaps in research teams whose foreign-born members were mid-process in HL3 adjudication at the time of the shutdown. At the City 17 hub, more than 200 candidates who had completed initial vetting and were awaiting final adjudication have been left in procedural limbo — unable to proceed under HL3 and not formally redirected to any alternative pathway.

"These are people who uprooted their lives," said immigration attorney Monica Salas, who represents a number of former HL3 applicants. "They turned down offers in Europe, in Canada, in places that wanted them. They made a bet on this country. And now they are just — stranded. Some of them describe it to me as feeling completely alien, suspended between worlds, not quite here and not quite anywhere else. They did everything right. The system changed on them."

Industry groups representing the semiconductor, biotechnology, and advanced manufacturing sectors have sent a joint letter to the White House warning that the shutdown will accelerate a talent migration to competitor nations. The letter, signed by more than 80 chief executives, warns of what it calls "cascading effects throughout the innovation ecosystem" — a resonance cascade, in the words of one signatory's communications director, that will be exceedingly difficult to reverse once it gathers momentum.

What the Critics Are Saying

The response from outside government has been swift and largely scathing.

"This is not complicated," said Ezra Klein, the journalist and author who has written extensively on the political economy of immigration and technological competitiveness. "We have a program. The program works. Independent auditors confirmed that the program works. And we terminated it based on a memo written by someone whose professional expertise is in shipping video games. I would genuinely like to know what problem, exactly, we have solved here. Because the problems we have created are very legible."

Bill Maher, the comedian and political commentator, addressed the shutdown on his podcast with characteristic bluntness. "I know this administration wants to be seen as tough on immigration. Fine. I understand the politics. But there is a difference between being tough and being stupid. This is a program for quantum physicists and oncology researchers and semiconductor engineers. You think China is shutting down its equivalent of this program? Because I promise you, China is not shutting down its equivalent of this program."

Don Lemon, appearing on a panel discussion that was widely shared on social media, framed the closure in more personal terms. "What this tells people — what it tells the next generation of the world's brightest minds — is that America's commitment to welcoming talent is conditional. That it can be reversed overnight, for reasons that have nothing to do with merit or security or economics. That is a reputational cost that does not appear in a DOGE spreadsheet. But it is very, very real."

Dr. Freeman

Among the program's most quietly regarded staff members was a junior researcher assigned to the City 17 hub's internal evaluation team. Gordon Freeman, 34, holds a doctorate in theoretical physics from the Massachusetts Institute of Technology and joined HL3's research division in 2023. Former colleagues describe him as exceptionally capable, deeply private, and constitutionally averse to institutional attention of any kind.

He was, by several accounts, among the most productive members of the evaluation team in the final year of the program's operation — someone who tended, colleagues say, to let his work speak entirely for itself.

Dr. Freeman did not respond to multiple requests for comment for this article.

The Longer View

Programs like HL3 are not easy to rebuild. The institutional knowledge embedded in its staff — the relationships with partner organizations, the refined vetting procedures, the hard-won interoperability with legacy federal systems — does not survive a termination notice. It disperses. Former staff find positions in other sectors or other agencies. The accumulated learning of a decade of immigration policy refinement, compressed into four years of intensive operation, does not transfer cleanly to an archive.

"There is a reason it is called a fast-track program," said Sandra Malik, the former deputy director. "Because speed matters. The competition does not pause while we deliberate. Every month that passes, every candidate we lose to a competitor nation, every researcher who decides the American process is no longer worth the uncertainty — that is not an abstraction. That is a human being, with a mind and a skill set, who is now building something for someone else."

The question of whether HL3 can be revived, in some form, under a future administration is one that policy experts are already discussing — quietly, contingently, with the particular caution of people who have watched a promising institution be dismantled before.

"It will take years," said one former program administrator, who has already begun fielding calls from former colleagues about what comes next. "And in those years, things will happen. Breakthroughs will happen somewhere else. Therapies will be developed somewhere else. Technologies will be commercialized somewhere else. Not because we lacked the capacity. Because we chose not to build it."

She paused, visibly exhausted.

"I have had a headache ever since the day that memo came out," she said quietly. "I am not sure it is going away any time soon."