The White House Office of Science and Technology Policy on Tuesday released a report called “Science: A New Golden Age,” arguing that President Donald Trump’s administration can overhaul US research policy for a new era of discovery.
The document matters because it sketches how Michael Kratsios, the OSTP director, wants federal science agencies to fund research, judge merit and work with industry. It lands while the same administration has proposed sharp cuts to science agencies, sought more power for political appointees to cancel grants awarded on scientific merit, and slowed the release of some congressionally approved research money, according to Ars Technica.
OSTP frames the report as an heir to “Science, the Endless Frontier,” the post-World War II policy argument for sustained government support of basic research. The new report says that earlier model remains important but needs substantial revision because AI, commercial technology and global competition have changed the conditions for science.
Less paperwork, less peer review
The report identifies a complaint familiar to working scientists: researchers spend too much time writing grant applications and handling administrative requirements. OSTP says agencies should simplify rules and streamline funding applications, though the report does not name specific regulations it wants changed.
It also argues that peer-review panels can block high-risk ideas by converging on consensus. OSTP proposes alternative funding models, including reviews by single decision-makers who may or may not be trained scientists. That would likely fund a wider range of ideas, but the report does not address the obvious failure mode: more money for weak or fringe proposals dressed up as boldness.
The document still endorses “gold standard science,” a Trump administration phrase that includes peer review in other contexts. That leaves the report in an awkward place: peer review is treated as both a quality marker and a bottleneck, depending on which policy point OSTP is making.
Innovation, commercialization and AI
OSTP says US science has become too incremental and should reward more transformative work. The report recommends new funding mechanisms and evaluations of whether grants produce the intended results. It does not explain how agencies should measure “innovation” in a way researchers would regard as objective.
The report also rejects a one-way model in which government funds basic science and companies later commercialize the results. OSTP favors a tighter loop among science, technology development and industry. The document points to the Human Genome Project as an example of government-led science, including its open data practices. Ars Technica noted a tension there: the genome project was international, while the Trump administration has pushed to restrict some international research collaborations, and open data can help foreign competitors as well as US companies.
AI receives a starring role. The report imagines a market of autonomous agents, bounties, internet-connected labs and cryptographically signed experimental results. That is a venture-capital flavored vision, unsurprising given Kratsios’ past work with Peter Thiel, but the report treats a still-provisional set of AI capabilities as a policy foundation.
A politics-heavy science agenda
The report criticizes diversity, equity and inclusion programs while also saying Americans with aptitude should have access to scientific training. It complains about foreign students taking doctoral slots, while Ars Technica noted that Trump administration funding attacks have already led some universities to reduce admissions.
OSTP also calls for more involvement from machinists, technicians and skilled craftspeople, including fellowships and practitioner-in-residence programs alongside Ph.D. researchers. The report invokes Detroit manufacturing as an example, though it does not clearly connect that history to a federal science research model.
Kratsios wrote that US researchers should one day feel free to pursue questions they previously could not. The report offers a long list of mechanisms: cut paperwork, weaken conventional peer review, involve industry earlier, lean on AI and bring hands-on workers into labs. It is less clear how those pieces fit with the administration’s funding cuts and grant cancellations, which are already shaping what scientists can pursue now.
This story draws on original reporting from Ars Technica.
