Trump’s AI “Manhattan Project” will fail if DOGE cuts are kept, critics say.
By executive order last month, Donald Trump launched his so-called “Genesis Mission.”
Described as a “historic national effort” to “invest in AI-enabled science to accelerate scientific advancement,” Trump claimed his mission would address key challenges to American energy dominance, innovation, and national security.
This mission, Trump boasted, would be a game-changer to science akin to putting a man on the moon or firing the first nuclear weapons. By building “an integrated AI platform” trained on “the world’s largest collection” of federal scientific data sets, he promised, the government could set off cascades of scientific breakthroughs.
Access to such a platform, Trump imagined, would supercharge top US labs, powering AI agents to do tasks like quickly test hypotheses and automate research workflows to speed up discoveries.
However, the mission crucially depends on strengthening collaboration between public, private, and academic sectors. And Trump’s order is concerningly vague on how those partnerships will be structured and funded at a time when many scientists have been sidelined due to a flurry of Trump orders earlier this year that eliminated their funding or removed them from their labs.
To critics, including scientists, policy experts, advocates, and historians, Trump’s order seems divorced from reality, given that he spent the past year attacking some of the very institutions the Genesis Mission would seem to depend on. Trump also seemed unclear about what can be achieved with AI and confused about how scientific progress is actually made, some critics suggested.
Among the critics was Arati Prabhakar, who served as the director of the White House Office of Science and Technology Policy under the Biden administration. Prabhakar told Ars that Trump’s crippling cuts to government science agencies, research grant funding freezes, and attacks on universities must be repaired or his mission will fail.
“After the Trump administration has inflicted so much damage to valuable datasets and publicly funded research, the new executive order is a Band-Aid on a giant gash,” Prabhakar said.
Also skeptical of Trump’s plans is Kathryn Kelley, executive director for the Coalition for Academic Scientific Computation, an educational nonprofit representing more than 100 of “the nation’s most forward-thinking universities and computing centers,” CASC’s LinkedIn said.
Her group is specifically dedicated to Genesis Mission-aligned goals, “advocating for the use of the most advanced computing technology to accelerate scientific discovery for national competitiveness, global security, and economic success.” And while Trump’s initiative could be considered “a step in the right direction,” Kelley told Ars that she shares Prabhakar’s concerns.
“Many research institutions and national laboratories continue to experience funding uncertainty, program disruptions, and workforce instability stemming from earlier cuts,” Kelley told Ars.
Particular pain points considered critical to address for Genesis Mission to move forward include reversing “impacts on staffing, ongoing research, and student pipelines,” she suggested.
In one prominent example, some Department of Government Efficiency (DOGE) cuts targeting workers at the National Science Foundation hit hardest in the branch designed to accelerate technology development across a wide range of research settings in the US. DOGE slashed workers there simply because it was the youngest directorate at NSF with the most workers in transition when Trump took office. As courts weighed legal challenges to cuts, whistleblowers warned that Trump was aiming to politicize and dismantle NSF.
“Large-scale initiatives like Genesis rely on highly skilled personnel, robust infrastructure, and sustained program support—some of the very resources at NSF and other federal agencies that were disrupted,” Kelley told Ars. “Rebuilding trust, re-establishing lost programs, and stabilizing the research workforce will be essential to make this mission feasible.”
Critics urged that Trump’s attacks on science also included messing with government datasets that scientists depend on. Since Trump’s second term started, scientists have watched valuable data get censored or scrubbed from government websites. Some researchers have rushed to recreate datasets independently with the help of the Internet Archive.
Prabhakar pointed out that some “datasets that could improve health and prevent disasters are eroding or even disappearing due to this administration,” while universities training “the next generation of great researchers and innovators have reduced or even stopped graduate admissions because of Trump’s assault.”
Without a massive undertaking to undo moves that critics have said undermined both US science and trust in it, Trump’s dreams of launching AI models that would propel a million moonshots could go down in history as merely hype.
“Without robust data and research and without people’s trust, America won’t lead in AI,” Prabhakar said.
For people in the science community, it’s hard to square Trump’s aggressive cuts from earlier this year with the broad ambition of Genesis Mission. Frustratingly, the president demands that scientists make discoveries on his timeline, without acknowledging AI’s limitations or how his attacks on science could be driving away talent that could help labs advance AI.
In many fields, scientists are still exploring how AI can aid research. Trump’s order appears to politicize science by focusing on areas he favors—like critical materials, nuclear energy, biotechnology, and quantum computing—despite their limited AI applications or data-quality challenges. Meanwhile, critics noted that it overlooks areas where “supercharging” AI could perhaps be more impactful—but where Trump notably does not want to leave his mark—like climate science or vaccine research.
It also lays out aggressive timelines for results, demanding that the Department of Energy Secretary, Chris Wright, “demonstrate an initial operating capability of the Platform for at least one of the national science and technology challenges” identified in less than a year (270 days). Ideally, Trump’s mission will have generated significant discoveries in key fields within the next three years before he leaves office, his order outlined.
Paul Josephson, a Colby College professor and expert in the history of 20th-century science and technology, told Ars that Genesis Mission deadlines differ from John F. Kennedy’s 10-year timeline to reach the moon.
Trump’s order “shows tremendous ignorance of how science and technology work,” Josephson said. The White House is saying, “Tell me what your discoveries will be and how many there will be in three years,” Josephson said, expecting that “we can pick the places where we want discoveries and make them happen.”
“That’s not anything like how science works,” Josephson told Ars, reducing Genesis Mission to “a vision without policy” and “a hope without funding.”
To Josephson, Trump’s order sounded “more like it came out of Silicon Valley” than out of talks with government scientists, seemingly rushing approvals of industry partnerships and incentives without mentioning what resources would be available to fund gutted labs or train the next generation of scientists. It’s perhaps notable that the order is DOE-centric and does not place the same emphasis on contributions from universities or national labs funded by NSF and the National Institutes of Health as it does on industry partners.
Kelley told Ars that “many public datasets are already being used effectively in research and industry” in the ways that Trump intends his AI platform to amplify. However, “there are areas—such as advanced nuclear research or emerging energy technologies—where datasets are limited.” And Trump risks reducing Genesis Mission to bluster by claiming that an AI platform could drive breakthroughs to the major challenges he flagged in the short term.
“There is a real risk that the EO’s ambitious framing could overpromise what AI can achieve in the near term without addressing foundational data gaps,” Kelley told Ars. “That said, even partial progress in these areas could provide valuable insights, but expectations need to be realistic.”
Just as important as asking where Genesis Mission funding is coming from or who the funding is going to, Chris R. Glass asks: “Where’s the talent coming from?”
Trump’s order does not forecast that, only vaguely referencing support for universities training scientists. This comes, of course, after the administration revoked an estimated $1.5 billion in federal grant money in 2025. Those grant cuts shrank the pipeline for PhD students at an “unprecedented rate,” Axios reported.
A Boston College professor who researches global student mobility and the impact of emergent technology on learning, Glass told Ars that Trump has notably left international talent out of his AI plans, despite the prominent roles that both “domestic and international scientists play in our current leadership” in AI.
In a recent Washington Post op-ed, Glass warned that “America is losing research scientists,” who are seeking more stable environments to set up their lives and conduct long-term studies.
As the Trump administration has attacked immigrants, other governments like the European Union and China have benefited by offering friendlier visa systems to attract the best and brightest minds graduating from US universities. Of course, of the two, China is America’s bigger AI rival. Earlier this year, China began heavily recruiting American scientists spooked by Trump’s grant funding cuts, and Glass confirmed that China has continued those efforts with the AI race heating up. Meanwhile, Trump appears to be going the other direction, recently requiring a $100,000 payment for some skilled workers seeking non-immigrant visas.
Throughout 2025, US universities’ ability to attract international students showed resilience, but “we’re on thin ice,” Glass told Ars, with that resilience “waning.”
Currently, the US “is ranked the lowest among top destinations for its safety and welcoming and the lowest for its post-graduation visa policies,” Glass said, noting that doctoral students must affirm that they do not intend to immigrate, even though the majority of STEM PhD students stay in the US after graduating.
“They want to stay, and we want them to stay,” Glass told Ars.
Another concerning outcome of Trump cuts that could hamper Genesis Mission: Entire research groups at many institutions were “displaced”—removed from their labs and left to work in cubicles without access to their equipment, Glass told Ars.
“I think scientists want to go where the best sciences are being done, but eventually these kinds of friction points and these hostile policies make them redirect elsewhere, even temporarily redirect, earn their doctorate in Europe and hope that the policy environment in the US changes,” Glass said.
To turn it around, Glass made several recommendations in his op-ed to help retain PhD graduates and create stable pathways for high-value talents. That includes suggesting that the Trump administration consider fast-tracking green cards for students in fields that Genesis Mission depends on, including AI and machine-learning researchers, quantum computing scientists, and semiconductor engineers.
He also thinks the US should “unlock the O-1A visa for researchers and entrepreneurs” by redefining what makes someone an “extraordinary” talent and creating dedicated “founder tracks” for international talent, as Britain and Singapore do. That visa is “uncapped yet underused,” Glass said, only approving 4,500 STEM candidates in 2023.
Without changes to the visa system, the US “risks redirecting those talent flows,” he said. “And like a river, once those talent flows get redirected, they are very difficult to reverse.”
And it won’t just be international talents jumping ship, Glass suggested, but also possibly US scientists forced to continue navigating potentially more of Trump’s cuts and indirect costs in the coming years.
“I think that’s the kind of thing that slowly eats away at someone’s desire to continue to do science in the United States,” Glass said.
Glass told Ars that he expects the US to stay on a “downward trajectory,” driving away talent in 2026, which Josephson suggested “will damage science both for the short and long term.”
“Many universities figured out a one-year contingency plan, but reality will set in if funding continues to be cut,” Glass said.
CASC’s Kelley told Ars that like university international student recruitment, “the US research ecosystem continues to be resilient, but the gap between ambitious goals and the current capacity must be carefully managed.”
“While the Genesis Mission signals strong intentions to invest in science and technology, its success will depend on aligning resources, rebuilding workforce capacity, and thoughtfully integrating AI and data capabilities where they are most effective,” Kelley said.
A scientist might be best positioned to understand the nuance that requires, but Josephson noted that Trump tasked his Science Advisor, Michael Kratsios, with leading the initiative. Unlike prior officials serving in that role, Kratsios is not a scientist and has no PhD, earning his BA in politics. Instead, Kratsios has strong industry ties, previously serving as chief of staff for venture capitalist Peter Thiel and managing director of a company called Scale AI.
To Josephson, Kratsios as head of the mission—which “seems to be totally based on faith in AI and datasets to do everything”—makes the initiative seem more aligned with Silicon Valley ambitions than public good. That could be a problem since historically, it has never worked when governments attempt to “pick winners” or pass industrial policy with claims that “if we do this, we will come out on top.”
“It’s a belief in AI as the cure or the panacea for all the world’s problems to ensure we’re a dominant technological power, but ignoring climate change, race, gender, anything that is important in daily life,” Josephson said.
Josephson is also an expert in Russian and Soviet history, explaining that precedent shows there are “tremendous dangers” of governments controlling which sciences are funded. In some ways, he thinks Genesis Mission “smells of Putin,” he told Ars, warning that Trump’s attempts to hoard and censor science in 2025 have been “as damaging to science and technology in the world’s leading centers as totalitarian regimes have been.”
“It reflects the general timbre of the Trump administration toward the scientific enterprise,” he suggested, saying that the president has embraced the “authoritarian view” that he “has the right to pick and choose which fields and which branches merit more support and which should not be funded at all.”
Jules Barbati-Dajches, an analyst for the Center for Science and Democracy at the Union of Concerned Scientists (UCS), told Ars that in addition to cuts, Trump recently “weakened federal agency policies (called scientific integrity policies) that were specifically in place to protect federal agency science from political interference.” This further threatens scientific integrity, Barbati-Dajches warned in August.
UCS has tracked “instances of science being sidelined, ignored, or misused by the federal government” across “multiple presidential administrations” for two decades, Barbati-Dajches told Ars. And although their methodology was recently updated, the current Trump administration stands out, as “the rate and impact of attacks on science we’ve observed over the past nine months far outpace anything UCS has tracked before,” Barbati-Dajches said.
Additionally, UCS has documented “cases of the administration using AI in their reports and research that raise concern” that AI initiatives like Genesis Mission may promote dubious claims to serve “politicized” outcomes, Barbati-Dajches said.
“This altogether paints a very troubling picture,” Barbati-Dajches said. “Scientific innovation and discovery are exciting, important, and can help inform federal policy and guidance. But as history tells us (and recent history even more so), science in the federal government needs protective guardrails to keep it independent and free from undue influence.”
With Trump pushing for rapid buildouts of AI data centers—sparking widespread backlash among Americans—Barbati-Dajches noted that UCS has documented his administration making “policy choices and decisions that benefit favored interests (including tech and fossil fuel companies) over the health and safety of the public and planet.” Genesis Mission appears to follow that trend, critics suggested, along with Trump’s most recent executive order threatening to block state AI laws, which many consider a gift to the tech industry.
