ADAPTING OUR INSTITUTIONS

Our task is to build a scientific enterprise that maximizes the marginal returns to intelligence, which means confronting, clearly, the factors that constrain its work. Three stand out. The first is the speed of feedback. Particle physicists have devised myriad theories about our universe, but we currently lack the experimental data to distinguish among them. No amount of intelligence can conjure observations that do not yet exist.

The second is the speed of atoms. Cells divide on their own schedule, and hardware must be physically manufactured. No cognitive power can alone make a rocket or a semiconductor fab build itself overnight.

The third is the constraint imposed by our own institutions. Publishing systems, funding structures, regulatory frameworks, clinical trial requirements, and cultural inertia govern how research happens. This will be felt first within the scientific enterprise. The coming abundance of cognitive capability demands a corresponding transformation of our scientific institutions. New technologies like AI should give us tremendous optimism, but we should harbor no illusions that AI will repair our scientific enterprise by default. Even if every scientist benefits from adopting AI, it does not follow that science as a whole will advance. This is one of the counterintuitive properties of complex systems: individual gains do not automatically aggregate into collective progress. For a century, the United States aggressively suppressed forest fires, and each intervention seemed like an obvious success. But by preventing small fires, we allowed the fuel of dead wood, dense brush, and dry undergrowth to accumulate until the inevitable fires became infernos even harder to contain. AI could do the same to science, by making each researcher seemingly more productive while allowing the conditions for catastrophic and systemic dysfunction to build.157 AI tools will make it trivial to generate more papers, more grant applications, and more submissions to peer review. But if the obstacle to scientific progress were simply the production of these artifacts, we would already be living in a scientific golden age. We are not.

As Chapter II documented, the exponential growth in publications, researchers, and funding over the past half-century has not produced a corresponding acceleration in discovery. Disruptive work represents an ever-shrinking fraction of total output, and the breakthroughs that reorient fields arrive no faster than they did generations ago.

The rise of AI in science therefore demands that we update the institutional machinery governing what gets funded, what gets published, what gets rewarded, and what gets corrected. Science works only if we can generate and verify knowledge in tandem. Models that make flawed paradigms easier to extend may entrench scientific biases; AI tools that flood peer review with unthoughtful submissions will overwhelm systems already stretched thin. In an age of abundant intelligence, everything discussed in the preceding chapters, such as the public-private partnerships that direct AI toward problems that matter, the metascience reforms that restore accountability and reward genuine exploration, the novel funding mechanisms that tolerate early failure, the rebuilding of our technical workforce and manufacturing base, becomes more urgent, not less.

BUILDING THE INFRASTRUCTURE

The reforms proposed in the preceding chapters are also essential for AIpowered science. AI has shifted basic research toward industry, made cross-disciplinary collaboration essential, and sharply increased the capital intensity of frontier science.

In doing so, AI has made the institutional adaptations described in the preceding chapters necessary to update the scientific enterprise for the modern world. The novel organizations and funding mechanisms in Chapter II matter because fully leveraging AI for science demands tight iteration between exploration and engineering, something traditional academic departments were not built to sustain. It also requires close partnerships between domain scientists who understand AI models and AI researchers who understand the science. Private research institutes are now housing machine learning researchers and life scientists in shared facilities to maximize collisions,158 while fellowship programs pair AI researchers with academic co-advisors.159 These reforms must spread across the entire scientific enterprise. Eventually, we will need AI-native scientific institutions, with rules, infrastructure, capital models, and cultural norms built around the use of powerful AI systems.

The public-private partnerships in Chapter III matter because frontier AI capabilities concentrate in private laboratories. The compute clusters required to train frontier models cost billions of dollars. The engineering teams capable of operating them mostly work at a handful of companies, on payrolls no university can currently match. Keeping private sector capabilities in conversation with the public scientific enterprise will require carefully crafted structures. Existing partnerships between federal laboratories and AI companies represent steps toward that future, with researchers applying cutting-edge AI systems to advance fusion energy, drive advancements in computational biology, and even control a rover on Mars.160

Finally, the reconnection of science and craft in Chapter IV matters because as intelligence becomes more abundant, the constraints on scientific progress shift from generating ideas to realizing them in the physical world. The world of atoms, which includes manufacturing, fabrication, and the infrastructure on which discovery depends, will increasingly determine the pace of progress. No amount of intelligence substitutes for the instruments needed to collect experimental data or the facilities needed to build and test new technologies. Reforms to rebuild apprenticeships, capture tacit knowledge, and reconnect universities with regional manufacturing ecosystems are focused precisely on this constraint.

THE GENESIS MISSION

Throughout our history, from the Manhattan Project to the Apollo Program, America’s greatest scientific advances have come when national capabilities were matched with the right institutional design. The Genesis Mission is the next chapter in that tradition, a national effort to harness AI for scientific discovery at a scale no other nation can match.

President Trump launched the Genesis Mission in November 2025 as America’s premier –“AI for science” initiative. The Executive Order establishing the Mission directs DOE to build the American Science and Security Platform, which will connect our most powerful supercomputers, AI systems, and scientific instruments and datasets into a single discovery engine designed to double the productivity and impact of American science and engineering within a decade.161 The Mission draws on an unparalleled base of national capability. DOE’s 17 national laboratories constitute the largest concentration of scientific infrastructure in the world, employing roughly 40,000 scientists, engineers, and technical staff, and receiving approximately $20 billion in annual funding.162 These institutions house our most advanced particle accelerators, synchrotron light sources, supercomputers, and experimental facilities spanning materials science, fusion energy, and nuclear security. Beyond our national laboratories, agencies including the FDA, NSF, National Oceanic and Atmospheric Administration (NOAA), and Department of Veterans Affairs steward vast quantities of scientific data accumulated over decades of federal investment, from genomic sequences to weather simulations.

The Genesis Mission will unlock this capacity, including by building foundational technologies and AI-ready datasets, to tackle the nation’s most complex science and technology challenges. Realizing the Mission’s potential requires addressing four key challenges that would otherwise constrain AI-enabled science. The first is problem selection. Not every scientific problem is well-suited to AI intervention. The strongest candidates exhibit certain characteristics, such as, for today’s AI systems, large combinatorial search spaces, substantial quantities of structured data, and clear metrics against which to benchmark progress. Protein structure prediction, for example, fit these criteria precisely.

The space of possible configurations is vast, decades of crystallographic data provided training material, and benchmarks such as the Critical Assessment of Protein Structure Prediction (CASP) allowed researchers to measure improvement. The Mission has therefore directed DOE to identify at least 20 science and technology challenges of national importance, spanning advanced manufacturing, biotechnology, critical materials, nuclear fission and fusion, quantum information science, and semiconductors. Challenges will be reviewed annually to reflect both scientific progress and national priorities. In a world where AI research is flush with capital, the Federal Government’s value-add is not funding AI in the abstract, but directing it toward problems where breakthroughs could unlock entire branches of downstream discovery and application, just as cracking the human genome did decades ago.

The second is institutional capacity. The Genesis Mission is designed to operationalize the reforms outlined throughout this report, many of which are preconditions for AI-powered science, at national scale. In December 2025, DOE announced agreements with twenty-four organizations, including leading AI companies, semiconductor manufacturers, and cloud providers.163 These partnerships, and the many that follow, will ensure the Mission’s outputs flow across the entire national research ecosystem. Furthermore, the Transformational AI Models Consortium, a cornerstone investment in the Mission, will mobilize National Laboratories to partner with industry to generate new AI-ready data and support the development of foundation models that harness DOE’s unique data, facilities, and expertise across scientific and engineering domains.164 The third is data infrastructure. Scientific data is the raw material for AI-powered discovery, but much of America’s most valuable data is currently inaccessible, uncurated, or locked behind restrictive licensing. Fixing this requires two complementary approaches. One is opening access to Federal Government data. Many valuable datasets exist because the government chose to build them, like NOAA’s weather data or the Materials Project’s mapping of inorganic crystals.165 The American Science Cloud, a cornerstone of the Mission’s infrastructure, will empower the National Labs to curate and distribute DOE’s AI-ready scientific data for the broader research community and unlock data hidden behind government bureaucracy. Approach two is creating incentives for individual researchers to curate and share their own datasets. Much valuable data arises organically, when individuals closest to the research recognize that their experimental records or failed trials could be helpful to others. This data is routinely abandoned, sometimes due to a lack of stable funding for storage and curation, and other times because there is no signal on the value of the information.166 Data on laboratory procedures and challenging experiments, for instance, may prove highly valuable in light of potential lab automation, yet such data is currently scattered. The Mission will address this directly, creating new funding opportunities for dataset curation and building new incentives to partake in these curation efforts across our science agencies. The fourth is the integration of AI capabilities with experimental infrastructure. Where materials discovery can take around 20 years from laboratory to 64 Chapter V – A New Golden Age deployment, closed-loop autonomous experimentation could collapse that timeline by an order of magnitude.167 That makes leadership in this platform technology a strategic imperative for the United States. We have already prototyped autonomous facilities across our national laboratories, such as the A-Lab at Lawrence Berkeley, which works on solid-state synthesis of inorganic materials, and the Polybot at Argonne, a modular robotics platform for materials characterization. But other countries, including Canada and China, are racing forward. The constraint on further automation runs deeper than funding. Decades of consolidation and offshoring in the scientific instruments industry have created pathologies one would expect, including expensive products, poor software, and proprietary data formats that lock researchers into vendor ecosystems. Scientists attempting to build automated workflows spend months simply getting different instruments to communicate. Scientific instruments themselves must be redesigned for automation from the ground up, with open interfaces and standardized data formats. The national laboratories deploy the largest concentration of advanced scientific instrumentation in the world, and their purchasing power can drive that redesign. The Mission has already invested in 14 projects focused on robotics, automated laboratories, and autonomous control of large-scale experiments.169 These efforts build on a parallel push at NSF to invest an initial $380 million into programmable cloud labs across academic institutions and startups, kicking off our domestic autonomous experimentation industry, just as NSFNET played an instrumental role in forming the backbone of the modern internet in the 1980s.170 America’s brightest minds and industries have always answered the call when their country needed them most. The Genesis Mission is that call to this generation of scientists and engineers to advance American scientific leadership in the era of AI. Its design reflects the core convictions of this report. The central role of the Federal Government is to shape the arena rather than direct discovery, recognizing that the private sector possesses capabilities public institutions must learn to leverage rather than replicate.

GOLD STANDARD SCIENCE

The promise of AI-enabled science rests on a foundation that is, at present, potentially unsound. We are preparing to train AI systems on scientific literature, deploy them to generate hypotheses, and trust them to design experiments, but the knowledge base they will draw on is riddled with error.

As detailed in Chapter II, a majority of researchers now acknowledge that science faces a reproducibility crisis. Between one-half and two-thirds of psychology studies failed replication attempts,171 more than one third of celebrated experimental economics studies similarly failed to replicate,172 and one study found that irreproducible findings in preclinical biomedical research alone misdirect an estimated $28 billion annually.173 Every false result can mislead subsequent researchers, creating cascading failures throughout the enterprise. Increased scientific productivity will not mean anything if the underlying findings are false. In May 2025, the President signed an executive order, Restoring Gold Standard Science, to begin addressing this dysfunction.174 The order establishes principles to govern all federally funded research: reproducibility; transparency; communication of error and uncertainty; collaboration across disciplines; skepticism of assumptions; falsifiability of hypotheses; unbiased peer review; acceptance of negative results; and freedom from conflicts of interest. Agencies are directed to apply an approach based on the weight of scientific evidence, transparently evaluating each piece of information based on quality and relevance. Replication does not happen at scale, in part, because of a market failure. Strong incentives drive researchers to publish new findings, with funding and prestige both flowing from novel claims. On the other hand, verification carries weak incentives; little glory comes from confirming someone else’s work. Previous attempts at large-scale replication have failed because they required armies of specialists to verify each study by hand. Manual verification cannot scale to the millions of papers published annually, and the problem is about to grow far more acute. As AI is introduced into the scientific process, it risks compounding these problems. False findings will multiply as it becomes easier to generate plausible-sounding scientific results than to verify them. AI research offers a preview. Leading conferences have seen submission surges of 60% in a single year, overwhelming the field’s capacity to evaluate new results. Researchers are now burdened with reviewing nonsensical AI-generated submissions while rebutting low-quality AI-generated reviews of their own work.175 Other fields will follow the same trajectory. The Genesis Mission is building a science generator with instruments capable of producing scientific discovery at an unprecedented scale. To sustain progress, we must also build its necessary counterpart: a verifier equal in rigor and scale. This is the central challenge that must be undertaken to address the reproducibility crisis and capture the full benefits of AI for science.

AI itself could help close the generation-verification gap, but only if we invest in the necessary infrastructure. AI has already begun to automate significant parts of the scientific workflow. Meanwhile, the Gold Standard Science requirements, including reproducibility, data sharing, and methodological documentation, create precisely the conditions under which automated verification becomes possible. The combination of both could lead to low-cost, continuous AI-enabled verification. Researchers have already outlined one vision of such a system, in which specialized agents parse submitted papers, reconstruct computational environments, execute analyses in sandboxed settings, and compare outputs against claimed results.176 The same infrastructure that audits human-authored papers today could tomorrow judge which machine-generated hypotheses merit experimental resources. Rising to this moment of need, NIH has launched a new, agency-wide initiative to elevate replication and reproducibility studies, identifying critical research and infrastructure needs to advance rigorous findings that are verifiable and transparently shared.177

Looking forward, the Federal Government must continue to lay the connective tissue between verification infrastructure and our scientific enterprise. This means establishing open APIs and interoperability standards that allow verification capabilities to plug into journal submission systems, grant reporting platforms, and private-sector AI research tools; standards for replication packages that ensure computational research arrives in machine-auditable form; and prizes for successfully replicating or disproving influential papers. The result should be a verification system that is not occasional but continuous, low-cost, and commensurate with the scale of discovery we are now capable of producing.

IDEAS ON THE HORIZON

The printing press transformed what could be written, who could read, and how knowledge accumulated. The research university created entirely new apparatuses for producing knowledge. The tools emerging today will do the same, enabling new forms of collaboration, new standards for verification, and new mechanisms for allocating attention and credit. Developments in mathematics already underway offer a clear glimpse of the transformative potential of AI paired with Gold Standard Science. In mathematics, checking a proof is often far easier than discovering one, an asymmetry in favor of verification that makes it a natural proving ground for the potential of AI and Gold Standard Science. With proof assistants, the challenge of proving a novel mathematical result reduces to the formalization of a theorem statement and the construction of a chain of arguments that the proof assistant accepts. Nevertheless, formalization has historically been too laborious to matter. Translating a single theorem into machine-checkable code could take months of painstaking work. The Liquid Tensor Experiment, a project to formalize a result in condensed mathematics in 2020, consumed nearly two years of effort from expert practitioners.178 Over the past two years, large language models have begun to make it possible for mathematicians to translate ordinary mathematical writing into these formal languages in real time. The acceleration has been striking. In early 2024, an ambitious project set out to formalize the Prime Number Theorem with a proof assistant. After 18 months and the collaboration of more than twenty people around the world, it had made intermediate progress but remained stuck on core difficulties in complex analysis.179 Then, in September 2025, a startup using AI completed the project in three weeks, spanning 1,100 formally verified theorems and definitions.180 Mathematical collaboration has traditionally relied on small, trust-based networks where participation depended on reputation and proximity. Formal verification replaces that model with one grounded in mathematical certainty, allowing collaboration to scale beyond personal trust. It has been suggested that mathematicians of the future may become architects of industrialized systems rather than solo artisans.181 The profession could grow to include orchestrators who design proof strategies, domain experts who contribute specialized knowledge, and skilled practitioners who direct AI tools. The mathematics we pursue will change as well. When AI handles computational drudgery, entire classes of problems become tractable, opening new scientific frontiers. Proof assistants and AI-enabled verification in mathematics represent a prototype of the Gold Standard Science tools that could propagate across disciplines. Wherever checking an answer is easier than finding one, AI stands to reorganize not just scientific discovery, but the social structures that govern who does it and how.

RETHINKING SCIENTIFIC PUBLICATION

The journal system was designed for a different era. When scientific journals emerged in the 17th century, they served perhaps hundreds of active researchers who corresponded by post. Today, there are nine million full-time researchers worldwide, publishing millions of articles across tens of thousands of journals.

The infrastructure of scientific communication has not kept pace with the scale of science itself, and AI will only widen the gap.182 The publication system’s structural problems go beyond scale. Journals create artificial scarcity, rewarding secrecy rather than open collaboration.

A small number of anonymous, unpaid reviewers who may have vested interests, limited expertise, or simply not enough time, determine what counts as legitimate science. The format rewards polished narratives over honest accounts of the research process. Null findings, failed experiments, methodological details, and the true rationale behind research choices rarely reach publication. These challenges will only sharpen with AI-enabled science. When anyone can generate plausible-looking research at industrial scale, the current metrics for evaluating scientific productivity, like papers published, citations accumulated, and impact factors achieved, will all fall to Goodhart’s Law as gameable targets. As information technology evolves, select research organizations backed by private funding have stopped supporting traditional journal publications. Their researchers release findings through alternative channels, including preprints, data repositories, and dynamic notebooks, which get reviewed and replicated rapidly within their community. They find that when researchers stop optimizing for publishable units, they design experiments differently. They become more creative, more collaborative. They care about whether results are useful rather than whether they make a compelling story.

The future of scientific communication may look very different from the present. Researchers might release shorter outputs more frequently, including datasets, code, preliminary findings, and methodological notes. Dynamic papers could update automatically as underlying data changes. Public peer review, conducted in the open rather than behind closed doors, could offer faster feedback loops. This is clearly seen by reference to machine learning communities, which already rapidly replicate papers posted to online repositories and turn social media platforms into forums for debate.

NEW FORMS OF COLLABORATION AND CREDIT

Today’s frontier advances in AI-for-science, such as AI models and autonomous laboratories, remain largely reflective of the existing structure of science. But combined with emerging decentralized technologies, they point toward the possibility of a more profound transformation in AI agents. Those agents would not merely assist human researchers, but participate as autonomous actors in a scientific economy. 69 Chapter V – A New Golden Age One key building block of this transformation will be more granular credit attribution. Blockchain-based systems can create immutable records of scientific contributions, timestamping every dataset uploaded, every analysis run, and every hypothesis proposed, and linking each to its creator.183 When the record is fully traceable and captures every contribution comprehensively, credit attribution need not be zero-sum. Contributions to shared resources, such as datasets, code libraries, and protocols, become properly visible and rewardable. Another building block will be new modes of financial transaction for scientific knowledge. Decentralized Autonomous Organizations, communities that pool resources and allocate them through collective governance, are beginning to fund scientific research directly without going through traditional institutional gatekeepers.184 Prediction polls, augmented with proper scoring feedback and statistical aggregation, have also been shown to forecast scientific developments better than prediction markets, based on technological trends already underway.185 Together, these mechanisms can direct resources toward problems based on the wisdom of crowds rather than committee review, potentially faster and more effectively. These pieces lay the foundation for a continuous, market-mediated, agentbased scientific economy. Imagine a funder posting a million-dollar bounty for the first validated therapeutic target for a rare disease. An agent working on adjacent problems notices a promising lead and posts a smaller bounty for replicating the finding. Other agents assess whether the problem falls within their competence, bid for the work, and contract an autonomous laboratory accessible through the internet, which runs the experiment and returns cryptographically signed results. The agent evaluates the evidence, updates its models, and publishes conclusions to a distributed ledger. When results prove ambiguous, human experts provide the judgment that automated systems lack. Smart contracts release funds automatically as milestones are verified. In such a world, experimental information becomes a tradeable commodity, and price mechanisms replace slow institutional coordination. Markets could form to support the scientific enterprise, such as prediction markets informing grantmakers about technologies on the horizon, bounty markets directing resources toward unsolved problems, and reputation markets tracking which agents produce reliable results. Agents would interact directly, exchanging data, hypotheses, and compute time through microtransactions. The whole system runs continuously, at speeds no human institution could match, but is guided by human judgment about which breakthroughs merit large bounties, and which questions require framing that machines cannot yet provide.

An agent-based scientific economy will reshape what science gets done. Agents might specialize in replication, profiting by verifying or falsifying claims that humans find too tedious to check. Others might focus on negative results, which journals refuse to publish but which hold real value for anyone exploring the same territory. Unconstrained by disciplinary boundaries, career incentives, or the limits of human attention, agents could pursue the questions that matter most, rather than the ones that yield publishable results. Cloud laboratories become the factories of this economy. Robotic facilities already exist that can synthesize molecules, run assays, and return results without human intervention. As these facilities proliferate and standardize interfaces, they become nodes in a network that any agent can access. An AI pursuing a hypothesis about protein folding could contract with a lab in Colorado, run crystallography experiments, receive results within hours, and integrate them into its next round of reasoning as it collaborates with humans in Boston. Physical experimentation, long the bottleneck of empirical science, becomes as accessible as computation. None of this exists today in a mature form, but the pieces are emerging separately. Whether they will combine into something like the system sketched here, or into something we cannot yet imagine, remains unknown. But the vision belongs in the same tradition as Bush’s original argument, that the frontier of scientific knowledge is open, expansive, and worth pushing into. The duty to keep pushing falls squarely on us.

AS WE MAY BUILD For millennia, scientific knowledge and technological progress were bounded by the cognitive faculties of the human mind. Knowledge, however collective in its making, had to fit inside the heads of individual thinkers, flow through human patterns of communication, and conform to the social technologies we invented to guide inquiry. That era is ending. Our civilization has been built on bronze and steel, substances we discovered and exploited, but did not design. The 21st century will be built on materials we engineer from first principles, metamaterials that bend light in ways nature never attempted, programmable matter that reconfigures on command, selfassembling structures that grow like living things but serve engineered purposes. The progression from the forge to the semiconductor fab took centuries; the progression from semiconductor fab to molecular assembler may take only decades. 71 Chapter V – A New Golden Age We may begin to engineer cells as precisely as we now engineer circuits, programming immune systems to hunt malignancies with complete specificity, shaping cell differentiation and tissue growth to repair damaged organs, and designing therapeutics atom by atom rather than discovering them by trial and error. If we get all this right, within a generation, the diseases that today kill millions—like cardiovascular failures, neurodegenerations, and cancers—may yield one by one to instruments we are now starting to build. The technological transformation is already underway. In the first year of the Trump Administration, more than a trillion dollars of investment commitments have been secured for advanced manufacturing infrastructure and for technology companies building in the physical world. The best minds of a generation are bent on breakthroughs in machine intelligence and its applications to science. New companies are created every day to discover new materials, design revolutionary drugs, build fusion power, and explore unsolved conjectures in mathematics. In parallel, a revival in the crafts has made advanced technology possible. Americans are grinding precision bearings to tolerances measured in millionths of an inch, polishing optics for surgical lasers and microscopes, spinning carbon nanofibers for spacecraft and medical implants, and growing semiconductor crystals of inhuman purity. The nation is rediscovering its capacity to build, grounded in the recognition that the frontier advances on two kinds of knowledge: the explicit, which can be written down and taught, and the tacit, which can only be learned through practice. America’s strength has always come from a culture that honors both science and craft, and keeps both open to all with the aptitude and interest to learn. To sustain this progress, we must invent new ways of doing science. Science is the pool of knowledge that underlies our technological pursuit. The science of the coming decades could produce knowledge that no single person fully grasps, verified by systems that no single person fully audits, yet more reliable than anything we have built before. Future infrastructure for discovery may harness trillions of AI agents running experiments, testing conjectures, and surfacing insights across every scientific domain, with human researchers setting directions, posing questions, integrating findings, and making the judgments that require wisdom rather than computation. We urgently need to begin preparing for this AI-enabled future, by building the institutions, incentive structures, and information systems that let us trust what we cannot individually comprehend and steer what we cannot fully predict. When Vannevar Bush wrote to President Roosevelt, the nation faced a choice, whether to continue the wartime mobilization of science, or let the momentum dissipate. We chose to build. The institutions that emerged gave America 72 Chapter V – A New Golden Age a half-century of scientific dominance that translated into security and prosperity. But they are no longer sufficient for the new frontier we face today. This report has described what must replace them: new partnerships that bridge discovery and production, new mechanisms that reward boldness over consensus, new infrastructure that reunites science with manufacturing and craft, and preparations for an AI-transformed era of scientific discovery. Our competitors understand this; they are building their own systems to capture this next era of science and technology, and to shape what it will be used for. The task, then, falls to our generation to design the institutions, standards, and capabilities that can guide a scientific enterprise larger, faster, and less individually comprehensible than any in history. In doing so, we will determine not only the future of American technological prowess, but also the trajectory of human knowledge itself. Rising to this challenge is vital if America is to continue to deliver prosperity and security to its people. 73 1 Vannevar Bush, Science, the Endless Frontier, 75th anniversary ed. (National Science Foundation, 2020), xiv. 2 Bush, Science, the Endless Frontier, 1. 3 U.S. Department of Agriculture, “A Look at Agricultural Productivity Growth in the United States, 1948-2017,” USDA Blog, March 5, 2020, https://www.usda.gov/about-usda/news/blog/ look-agricultural-productivity-growth-united-states-1948-2017. 4 Elizabeth Arias et al., “United States Life Tables, 2023,” National Vital Statistics Reports 74, no. 6 (National Center for Health Statistics, July 15, 2025), https://www.cdc.gov/nchs/data/nvsr/ nvsr74/nvsr74-06.pdf. 5 Ching-Hon Pui and William E. Evans, “A 50-Year Journey to Cure Childhood Acute Lymphoblastic Leukemia,” Seminars in Hematology 50, no. 3 (2013): 185–196, https://pmc.ncbi.nlm.nih.gov/ articles/PMC3771494. 6 Earl S. Ford et al., “Explaining the Decrease in U.S. Deaths from Coronary Disease, 1980–2000,” New England Journal of Medicine 356, no. 23 (2007): 2388–2398, https://www.nejm.org/doi/ full/10.1056/NEJMsa053935. 7 Keith Fuglie et al., Agricultural Research and Development: Public and Private Investments Under Alternative Markets and Institutions, AER-735 (U.S. Department of Agriculture, Economic Research Service, May 1996), https://www.ers.usda.gov/publications/pub-details?pubid=40696. 8 Bush, Science, the Endless Frontier, 9. 9 Bush, Science, the Endless Frontier, 13. 10 Bush, Science, the Endless Frontier, xiii. 11 Bush, Science, the Endless Frontier, xiii. 12 Semiconductor Industry Association, 2025 SIA Factbook, https://www.semiconductors.org/ wp-content/uploads/2025/05/2025-SIA-Factbook-FINAL-1.pdf. 13 National Center for Science and Engineering Statistics, National Patterns of R&D Resources: 2023-24 Data Update, NSF 26-313 (National Science Foundation, February 2026), https://ncses. nsf.gov/pubs/nsf26313. 14 National Center for Science and Engineering Statistics, National Patterns of R&D Resources. 15 Bush, Science, the Endless Frontier, 17–21. 16 Donald E. Stokes, Pasteur’s Quadrant: Basic Science and Technological Innovation (Brookings Institution Press, 1997). 17 Sandra L. Schneider et al., 2018 Faculty Workload Survey: Primary Report (Federal Demonstration Partnership, 2020), https://thefdp.org/wp-content/uploads/FDP-FWS-2018-PrimaryReport.pdf. End Notes 74 End Notes 18 Pierre Azoulay et al., “Indirect Cost Recovery in U.S. Innovation Policy: History, Evidence, and Avenues for Reform” (NBER Working Paper No. 33627, National Bureau of Economic Research, June 2025), https://doi.org/10.3386/w33627; Congressional Research Service, “NIH Indirect Costs Policy for Research Grants: Recent Developments,” CRS Insight IN12516, April 17, 2026, https://www.congress.gov/crs-product/IN12516. 19 National Institutes of Health, “Supplemental Guidance to the 2024 NIH Grants Policy Statement: Indirect Cost Rates,” NOT-OD-25-068, February 7, 2025, https://grants.nih.gov/grants/ guide/notice-files/NOT-OD-25-068.html. 20 National Science Board, Discovery: R&D Activity and Research Publications, NSB-2025-7 (National Science Foundation, National Center for Science and Engineering Statistics, July 23, 2025), https://ncses.nsf.gov/pubs/nsb20257; Central Intelligence Agency, “A Comparison of Soviet and U.S. Gross National Products, 1960-83,” research paper, released as sanitized, 1999, https://www.cia.gov/readingroom/docs/DOC_0000498181.pdf. 21 National Science Board, Discovery: R&D Activity and Research Publications; Organisation for Economic Co-operation and Development, Main Science and Technology Indicators (OECD, 2026), https://www.oecd.org/en/data/datasets/main-science-and-technology-indicators.html. 22 National Center for Science and Engineering Statistics, Doctorate Recipients from U.S. Universities: 2023, NSF 25-300 (National Science Foundation, December 2, 2024), Figure 8, https://ncses. nsf.gov/pubs/nsf25300. 23 Donald Trump, National Security Presidential Memorandum 33: United States Government Supported Research and Development National Security Policy, January 14, 2021. 24 Chuck Gwyn and Stefan Wurm, “EUV LLC: An Historical Perspective,” in EUV Lithography, ed. Vivek Bakshi (SPIE Press, December 10, 2008), https://doi.org/10.1117/3.769214. 25 National Center for Science and Engineering Statistics, “Table 6-3: Temporary Visa Holder Research Doctorate Recipients with Definite Postgraduation Commitments, by Major Field of Doctorate: 2024” in Doctorate Recipients from U.S. Universities: 2024 Data Tables, NSF 25-349 (U.S. National Science Foundation, 2025), https://ncses.nsf.gov/pubs/nsf25349. 26 Bush, Science, the Endless Frontier, 17. 27 Andrew Fieldhouse and Karel Mertens, “The Returns to Government R&D: Evidence from U.S. Appropriations Shocks,” Working Paper No. 2305 (Federal Reserve Bank of Dallas, 2024), https://www.dallasfed.org/research/papers/2023/wp2305. 28 Nicholas Bloom et al., “Are Ideas Getting Harder to Find?” American Economic Review 110, no. 4 (2020): 1104–44; Michael Park et al., “Papers and Patents Are Becoming Less Disruptive over Time,” Nature 613 (2023): 138–44. 29 Jack W. Scannell et al., “Diagnosing the Decline in Pharmaceutical R&D Efficiency,” Nature Reviews Drug Discovery 11, no. 3 (2012): 191–200, https://doi.org/10.1038/nrd3681. 30 The NIH budget doubled from $13.6 billion in 1998 to$ 27.1 billion in 2003. See: Bruce Alberts et al., “Rescuing US Biomedical Research from Its Systemic Flaws,” Proceedings of the National Academy of Sciences 111, no. 16 (April 14, 2014): 5773–5777, https://doi.org/10.1073/pnas.1404402111. 31 Bloom, “Are Ideas Getting Harder to Find?” 32 Adam Mastroianni, “Ideas Aren’t Getting Harder to Find and Anyone Who Tells You Otherwise Is a Coward and I Will Fight Them,” Experimental History (blog), May 17, 2022, https://www.experimental-history.com/p/ideas-arent-getting-harder-to-find. 33 Paul Starr, The Social Transformation of American Medicine: The Rise of a Sovereign Profession and the Making of a Vast Industry, updated ed. (Basic Books, 2017). 34 National Institute of Allergy and Infectious Diseases, “Timeline for Funding Decisions,” National Institutes of Health, last reviewed September 30, 2024, https://www.niaid.nih.gov/grants-contracts/timelines-funding-decisions. 75 End Notes 35 Joe Sutter and Jay Spenser, 747: Creating the World’s First Jumbo Jet and Other Adventures from a Life in Aviation (Smithsonian Books, 2006). 36 Council on Governmental Relations, “Changes in Federal Research Requirements Since 1991,” January 2025, https://www.cogr.edu/sites/default/files/RegChangesSince1991_Dec%202024.pdf. 37 Pierre Azoulay et al., “Indirect Cost Recovery and American Innovation: Context and Ideas for Reform,” Institute for Progress, July 24, 2025, https://ifp.org/indirect-cost-recovery-andamerican-innovation. 38 Kristin R. W. Matthews et al., “The Aging of Biomedical Research in the United States,” PLOS One 6, no. 12 (2011): e29738, https://doi.org/10.1371/journal.pone.0029738. 39 National Institutes of Health, “Average Age and Degree of NIH R01-Equivalent First-Time Awardees 1980–2016,” Early Stage Investigator Related Data, NIH Grants and Funding, last updated September 9, 2024, https://grants.nih.gov/policy-and-compliance/policy-topics/ early-stage-investigators/related-data. 40 Max Planck, Scientific Autobiography and Other Papers, translated by Frank Gaynor (Williams & Norgate Ltd., 1950), 33–34. 41 Pierre Azoulay et al., “Does Science Advance One Funeral at a Time?” American Economic Review 109, no. 8 (2019): 2889–2920, https://doi.org/10.1257/aer.20161574. 42 Johan S. G. Chu and James A. Evans, “Slowed Canonical Progress in Large Fields of Science,” Proceedings of the National Academy of Sciences 118, no. 41 (2021): e2021636118, https://doi. org/10.1073/pnas.2021636118. 43 Ted Cruz, D.E.I.: Division, Extremism, Ideology: How the Biden-Harris NSF Politicized Science (U.S. Senate Committee on Commerce, Science, and Transportation, October 2024), https:// www.govinfo.gov/app/details/GOVPUB-Y4_C73_7-PURL-gpo234941. 44 Igor R. Efimov et al., “Politicizing Science Funding Undermines Public Trust in Science, Academic Freedom, and the Unbiased Generation of Knowledge,” Frontiers in Research Metrics and Analytics 9 (2024): 1418065, https://doi.org/10.3389/frma.2024.1418065. 45 Jerome Karabel, The Chosen: The Hidden History of Admission and Exclusion at Harvard, Yale, and Princeton (Houghton Mifflin, 2005). 46 Pierre Azoulay et al., “Incentives and Creativity: Evidence from the Academic Life Sciences,” RAND Journal of Economics 42, no. 3 (September 12, 2011): 527–554, https://doi.org/10.1111/ j.1756-2171.2011.00140.x. 47 Michael Park et al., “Papers and Patents Are Becoming Less Disruptive over Time.” 48 Open Science Collaboration, “Estimating the Reproducibility of Psychological Science,” Science 349, no. 6251 (2015): aac4716, https://doi.org/10.1126/science.aac4716. 49 Anatoly Nikolaev et al., “Retracted Article: APP Binds DR6 to Trigger Axon Pruning and Neuron Death via Distinct Caspases,” Nature 457 (2009): 981–989, https://doi.org/10.1038/nature07767. 50 Genentech, “Findings of 2023 Genentech Review of 2009 Nature Paper and Related Research,” April 6, 2023, https://www.gene.com/media/statements/ps_040623. 51 Nikolaev et al., “Retracted Article: APP Binds DR6 to Trigger Axon Pruning and Neuron Death via Distinct Caspases.” 52 Theo Baker, “Stanford President’s Research Under Investigation for Scientific Misconduct, University Admits ‘Mistakes’,” The Stanford Daily, November 29, 2022, https://stanforddaily. com/2022/11/29/stanford-presidents-research-under-investigation-for-scientific-misconduct-university-admits-mistakes; Retraction Note, Nature 625, no. 7993 (2024): 204, https://doi. org/10.1038/s41586-023-06943-3. 76 End Notes 53 Dalmeet Singh Chawla, “‘Golden Tickets’ on the Cards for NSF Grant Reviewers,” Nature 614 (2023): 604–605, https://doi.org/10.1038/d41586-023-00579-z; Villum Foundation, “The Villum Experiment,” https://villumfonden.dk/en/group/grantsubarea/villum-experiment. 54 R.D. Anderson, “Germany and the Humboldtian Model,” in European Universities from the Enlightenment to 1914 (Oxford University Press, 2004), https://doi.org/10.1093/acprof: oso/9780198206606.003.0004. 55 Nobel Prize Outreach, “The Nobel Prize in Chemistry 2024,” press release, October 9, 2024, https://www.nobelprize.org/prizes/chemistry/2024/press-release. 56 “The Audacious Project: Launching the Protein Design Revolution,” UW Medicine, April 17, 2019, https://give.uwmedicine.org/ipd-audacious. 57 John Moult et al., “A Large-Scale Experiment to Assess Protein Structure Prediction Methods,” Proteins: Structure, Function, and Bioinformatics 23, no. 3 (November 1995): ii–v, https://doi. org/10.1002/prot.340230303; “Protein Data Bank: Key to the Molecules of Life,” NSF Impacts, U.S. National Science Foundation, https://www.nsf.gov/impacts/protein-data-bank. 58 Michael Nielsen and Kanjun Qiu, “A Vision of Metascience: An Engine of Improvement for the Social Processes of Science,” October 18, 2022, https://scienceplusplus.org/metascience. 59 Ben Southwood, “The Rise and Fall of the Industrial R&D Lab,” Works in Progress, August 28, 2020, https://worksinprogress.co/issue/the-rise-and-fall-of-the-american-rd-lab. 60 Sam Rodriques and Adam Marblestone, “Focused Research Organizations to Accelerate Science, Technology, and Medicine,” Federation of American Scientists, September 24, 2020, https://fas.org/publication/focused-research-organizations-to-accelerate-science-technology-and-medicine. 61 Ben Reinhardt, “Unbundling the University,” Speculative Technologies, February 2025, https:// www.unbundle-the-university.com. 62 Caleb Watney, “Launching X-Labs for Transformative Science Funding,” in The Techno-Industrial Policy Playbook (The Foundation for American Innovation; American Compass; The Institute for Progress; New American Industrial Alliance Foundation, 2025), https://www.rebuilding.tech/ posts/launching-x-labs-for-transformative-science-funding. 63 Adam Marblestone et al., “Unblock Research Bottlenecks with Non-Profit Start-Ups,” Nature 601, no. 7892 (January 11, 2022): 188–190, https://doi.org/10.1038/d41586-022-00018-5. 64 Caleb Watney, “Launching X-Labs for Transformative Science Funding.” 65 Stuart Buck, “A Taxonomy of R&D Orgs: What Is New, What Is Missing?” Good Science Project, forthcoming. 66 Adam Marblestone and Andrew Payne, “Mapping the Brain for Alignment,” Institute for Progress, August 11, 2025, https://ifp.org/mapping-the-brain-for-alignment. 67 Heidi Williams, “Building a Better NIH,” Institute for Progress, May 17, 2023, https://ifp.org/ building-a-better-nih. 68 Chawla, “‘Golden Tickets’ on the Cards for NSF Grant Reviewers.” 69 Ishan Sharma et al., “Piloting and Evaluating NSF Science Lottery Grants: A Roadmap to Improving Research Funding Efficiencies and Proposal Diversity,” Institute for Progress, February 2, 2022, https://ifp.org/piloting-and-evaluating-nsf-science-lottery-grants. 70 Chawla, “‘Golden Tickets’ on the Cards for NSF Grant Reviewers.” 71 Howard Hughes Medical Institute, “Mid-Career & Senior Faculty Program,” https://www.hhmi. org/programs/investigators. 72 Pierre Azoulay et al., “Incentives and Creativity: Evidence from the Academic Life Sciences.” 77 End Notes 73 Francis S. Collins et al., “NIH Roadmap/Common Fund at 10 Years,” Science 345, no. 6194 (2014): 274–276, https://doi.org/10.1126/science.1255860. 74 NSF Graduate Research Fellowship Program, “About GRFP,” https://www.nsfgrfp.org/about. html. 75 Patrick Collison et al., “What We Learned Doing Fast Grants,” Future, June 2, 2022, https:// future.com/what-we-learned-doing-fast-grants; Heidi Williams, “To Speed Scientific Progress, Do Away with Funding Delays,” The Washington Post, August 14, 2023, https://www.washingtonpost.com/opinions/2023/08/14/heidi-williams-science-research-funding. 76 Defense Advanced Research Projects Agency, “The DARPA Grand Challenge: Ten Years Later,” DARPA News, March 13, 2014, https://www.darpa.mil/news/2014/grand-challenge-ten-yearslater. 77 XPRIZE Foundation, “Mojave Aerospace Ventures Wins That Competition That Started It All,” https://www.xprize.org/news/mojave-aerospace-ventures-wins-the-competition. 78 Vesuvius Challenge, “Vesuvius Challenge 2023 Grand Prize Awarded: We Can Read the Scrolls!” February 4, 2024, https://scrollprize.org/grandprize. 79 Vitalik Buterin et al., “A Flexible Design for Funding Public Goods,” Management Science 65, no. 11 (July 2, 2019): 5171–5187, https://doi.org/10.1287/mnsc.2019.3337. 80 Adam Marblestone et al., “Introducing the Convergent Research Gap Map,” Essential Technology (blog), April 15, 2025, https://www.essentialtechnology.blog/p/introducing-the-convergent-research. 81 “Heilmeier Catechism,” Defense Advanced Research Projects Agency (DARPA), https://www. darpa.mil/about/heilmeier-catechism. 82 UK Metascience Unit, A Year in Metascience (2025), (Department for Science, Innovation and Technology and UK Research and Innovation, June 30, 2025), https://www.gov.uk/government/ publications/a-year-in-metascience-2025. 83 The White House, National Security Strategy of the United States of America (November 2025), https://www.whitehouse.gov/wp-content/uploads/2025/12/2025-National-Security-Strategy.pdf. 84 Allan Dafoe, “On Technological Determinism: A Typology, Scope Conditions, and a Mechanism,” Science, Technology, & Human Values 40, no. 6 (2015): 1047–1076, https://doi.org/10.1177/ 0162243915579283. 85 W. Brian Arthur, Increasing Returns and Path Dependence in the Economy (University of Michigan Press, 1994). 86 National Science Board, Discovery: R&D Activity and Research Publications. 87 KPMG Private Enterprise, “2024 Global VC Investment Rises to $368 Billion as Investor Interest in AI Soars While IPO Optimism Grows For 2025 According to KPMG Private Enterprise’s Venture Pulse,” press release for Venture Pulse, January 2025, https://kpmg.com/xx/en/media/ press-releases/2025/01/2024-global-vc-investment-rises-to-368-billion-dollars.html. 88 U.S. Nuclear Regulatory Commission, “Pre-application Process,” last reviewed or updated May 13, 2026, https://www.nrc.gov/reactors/new-reactors/advanced/new-app/general-guidance/ pre-app-process; Office of Nuclear Energy, “NRC Approves First U.S. Small Modular Reactor Design,” U.S. Department of Energy, September 2, 2020, https://www.energy.gov/ne/articles/ nrc-approves-first-us-small-modular-reactor-design; Office of Nuclear Energy, “NRC Certifies First U.S. Small Modular Reactor Design,” U.S. Department of Energy, January 20, 2023, https:// www.energy.gov/ne/articles/nrc-certifies-first-us-small-modular-reactor-design. 89 Nuclear Energy Innovation and Modernization Act, Pub. L. No. 115–439, 132 Stat. 5565 (2019). 78 End Notes 90 Erik Cothron, “Fact Sheet: President Trump’s Nuclear Energy Executive Orders,” Nuclear Innovation Alliance, May 29, 2025, https://nuclearinnovationalliance.org/fact-sheet-president-trumps-nuclear-energy-executive-orders. 91 “PRIMA Visual Prosthesis,” Science Corporation, https://science.xyz/technologies/prima. 92 Aylin Sertkaya et al., U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation, Examination of Clinical Trial Costs and Barriers for Drug Development, July 24, 2014, https://aspe.hhs.gov/reports/examination-clinical-trial-costs-barriers-drug-development-0. 93 U.S. Food and Drug Administration, “FDA Eliminates Major Barrier to Using Real-World Evidence in Drug and Device Application Reviews,” press announcement, December 15, 2025, https://www.fda.gov/news-events/press-announcements/fda-eliminates-major-barrierusing-real-world-evidence-drug-and-device-application-reviews. 94 Vinay Prasad and Martin A. Makary, “One Pivotal Trial, the New Default Option for FDA Approval—Ending the Two-Trial Dogma,” New England Journal of Medicine 394, no. 8 (February 18, 2026): 815–17, https://doi.org/10.1056/NEJMsb2517623. 95 National Institutes of Health, “Highlighted Topics,” NIH Grants & Funding, last updated December 10, 2025, https://grants.nih.gov/funding/find-a-fit-for-your-research/highlighted-topics; National Institutes of Health, “Updated Application Policies: NIH Administrative Burden Reduction Effort Removal of Requirements for Letters of Intent and Unsolicited Applications Requesting $ 500,000 or More in Direct Costs,” notice no. NOT-OD-26-019, NIH Guide for Grants and Contracts, December 3, 2025, https://grants.nih.gov/grants/guide/notice-files/ NOT-OD-26-019.html. 96 U.S. Food and Drug Administration, “Commissioner’s National Priority Voucher (CNPV) Pilot Program,” last modified April 28, 2026, https://www.fda.gov/industry/commissioners-national-priority-voucher-cnpv-pilot-program. 97 Kim Stringfellow, “Peace, Love and Rockets: Amateur Rocketry in the Mojave,” PBS SoCal, August 31, 2017, https://www.pbssocal.org/shows/artbound/peace-love-and-rockets-amateur-rocketry-in-the-mojave; Friends of Amateur Rocketry, Inc., https://friendsofamateurrocketry.org. 98 Mojave Air and Space Port, “America’s First Inland Space Port,” https://mojaveairport.com/ about-us/page/americas-first-inland-space-port. 99 NASA, “NASA and X Prize Announce Winners of Lunar Lander Challenge,” press release, June 6, 2013, https://www.nasa.gov/news-release/nasa-and-x-prize-announce-winners-oflunar-lander-challenge. 100 NASA, “NASA Stennis Inks Expanded Test Complex Agreement with Relativity Space,” press release, September 7, 2023, https://www.nasa.gov/news-release/nasa-stennis-inks-expandedtest-complex-agreement-with-relativity-space. 101 “Office of Science User Facilities,” U.S. Department of Energy, https://www.energy.gov/science/ office-science-user-facilities. 102 National Nanotechnology Coordinated Infrastructure, https://nnci.net. 103 BioLabs, https://www.biolabs.io. 104 U.S. Food and Drug Administration, “Current Good Manufacturing Practice (CGMP) Regulations,” last modified November 21, 2025, https://www.fda.gov/drugs/pharmaceutical-quality-resources/current-good-manufacturing-practice-cgmp-regulations. 105 National Center for Science and Engineering Statistics, National Patterns of R&D Resources. 106 Foundation for the National Institutes of Health, https://fnih.org. 107 National Institutes of Health, “Accelerating Medicines Partnership (AMP),” last reviewed July 16, 2025, https://www.nih.gov/amp. 79 End Notes 108 “A Doctoral Program for Full-time Employees,” Northeastern University, https://phd.northeastern.edu/industry-and-experiential-phd-program. 109 “Activate,” Entrepreneur Futures Network, https://entrepreneurfutures.org/activate. 110 Robert Cook-Deegan, The Gene Wars: Science, Politics, and the Human Genome (W. W. Norton, 1994), 102. 111 Francis S. Collins et al., “The Human Genome Project: Lessons from Large-Scale Biology,” Science 300, no. 5617 (April 11, 2003): 286–90, https://doi.org/10.1126/science.1084564. 112 Simon Tripp and Martin Grueber, Economic Impact of the Human Genome Project (Battelle Memorial Institute, Technology Partnership Practice, May 2011), https://battelle.org/docs/ default-source/misc/battelle-2011-misc-economic-impact-human-genome-project.pdf. 113 United States General Accounting Office, Federal Research: SEMATECH’s Efforts to Strengthen the U.S. Semiconductor Industry, RCED-90-236, September 13, 1990, https://gao.gov/assets/ rced-90-236.pdf. 114 Sandia National Laboratories, “Partners Unveil First Extreme Ultraviolet Chip-Making Machine,” April 11, 2001, https://newsreleases.sandia.gov/partners-unveil-first-extreme-ultravioletchip-making-machine. 115 Gwyn and Wurm, “EUV LLC: An Historical Perspective.” 116 Alexis de Tocqueville, Democracy in America, trans. Henry Reeve, vol. 2, bk. 3, ch.18. 117 “Autonomous Vehicles Testing and Operating in the State of Arizona,” Arizona Department of Transportation, Motor Vehicle Division, https://azdot.gov/mvd/services/professional-services/ autonomous-vehicles-testing-and-operating-state-arizona. 118 “TSMC Arizona,” Taiwan Semiconductor Manufacturing Company, https://www.tsmc.com/ static/abouttsmcaz/index.htm. 119 Regulatory Sandbox Program Amendments, H.B. 217, 2021 Gen. Sess. (Utah 2021). 120 James R. Holbein and Justin Holbein, “Wyoming Laws More Crypto-Friendly with Issuance of State Stablecoin,” Braumiller Law Group, PLLC, https://www.braumillerlaw.com/wyominglaws-more-crypto-friendly-with-issuance-of-state-stablecoin. 121 Dan Wang, “How Technology Grows (a Restatement of Definite Optimism),” blog, July 24, 2018, https://danwang.co/how-technology-grows; Dan Wang, Breakneck: China’s Quest to Engineer the Future (W. W. Norton & Company, 2025). 122 Michael Polanyi, The Tacit Dimension (Doubleday, 1966). 123 Harry Collins, Changing Order: Replication and Induction in Scientific Practice (University of Chicago Press, 1992). 124 Wang, “How Technology Grows.” 125 Katelynn Harris, “Forty Years of Falling Manufacturing Employment,” Beyond the Numbers 9, no. 16 (November 2020), https://www.bls.gov/opub/btn/volume-9/forty-years-of-falling-manufacturing-employment.htm. 126 U.S. Bureau of Labor Statistics, “All Employees, Manufacturing [MANEMP],” FRED, Federal Reserve Bank of St. Louis, March 2, 2026, https://fred.stlouisfed.org/series/MANEMP. 127 U.S. Bureau of Labor Statistics, “All Employees, Manufacturing [MANEMP].” 128 Tim Bajarin, “Maker Faire: Why the Maker Movement Is Important to America’s Future,” Time, May 19, 2014, https://time.com/104210/maker-faire-maker-movement. 129 Will Holman, “Makerspace: Towards a New Civic Infrastructure,” Places Journal, November 2015, https://placesjournal.org/article/makerspace-towards-a-new-civic-infrastructure. 80 End Notes 130 Matthew B. Crawford, “Shop Class as Soulcraft,” The New Atlantis, no. 13 (Summer 2006): 7–24, https://www.thenewatlantis.com/wp-content/uploads/legacy-pdfs/TNA13-Crawford.pdf. 131 Nick Moore, “Connecting Talent to Opportunity: A National Challenge to Build Talent Marketplaces,” Homeroom Blog, U.S. Department of Education, January 13, 2026, https://www.ed.gov/ about/homeroom-blog/connecting-talent-opportunity-national-challenge-build-talent-marketplaces; Heather Hennerich, “The Jobs and Degrees Underemployed College Graduates Have,” Open Vault Blog, Federal Reserve Bank of St. Louis, August 13, 2025, https://www.stlouisfed.org/ open-vault/2025/aug/jobs-degrees-underemployed-college-graduates-have. 132 Rita R. Zota, “A Snapshot of Federal Student Loan Debt,” CRS Report No. IF10158, February 19, 2025, https://www.congress.gov/crs-product/IF10158. 133 Rainer Weiss, “Biographical,” NobelPrize.org, https://www.nobelprize.org/prizes/physics/2017/ weiss/biographical; Bryan Marquard, “Rainer Weiss, Nobel Prize-Winner Who Helped Unlock Secrets of the Universe, Dies at 92,” The Boston Globe, August 31, 2025, https://www.bostonglobe.com/2025/08/31/metro/mit-nobel-prize-winner-rainer-weiss-passes-away. 134 American Association of Community Colleges, “AACC Fast Facts 2025,” February 2025, https://www.aacc.nche.edu/wp-content/uploads/2025/02/AACC2025_Fact_Sheet.pdf. 135 U.S. Department of Labor, “US Department of Labor Announces $145M in Funds Supporting Performance-Based Registered Apprenticeship Expansion across Key Industry Sectors,” news release, February 13, 2026, https://www.dol.gov/newsroom/releases/eta/eta20260213-0. 136 Robert I. Lerman, “Proposal 7: Expanding Apprenticeship Opportunities in the United States” (The Hamilton Project, Brookings Institution, 2014), https://www.hamiltonproject.org/assets/ legacy/files/downloads_and_links/expand_apprenticeship_opportunities_united_states_lerman.pdf; Amy Simon, “Revitalizing the Federal Apprenticeship System,” American Compass, June 8, 2022, https://americancompass.org/revitalizing-the-federal-apprenticeship-system. 137 Steven Klepper, “Disagreements, Spinoffs, and the Evolution of Detroit as the Capital of the U.S. Automobile Industry,” Management Science 53, no. 4 (2007): 616–631, https://doi.org/10.1287/ mnsc.1060.0683. 138 Philip E. Auerswald and Lewis M. Branscomb, “Valleys of Death and Darwinian Seas: Financing the Invention to Innovation Transition in the United States,” The Journal of Technology Transfer 28 (2003): 227–239, https://doi.org/10.1023/A:1024980525678. 139 J.D. Vance, “Remarks at the American Dynamism Summit” (speech, Third American Dynamism Summit, Washington, D.C., March 18, 2025). 140 Tyler Buchanan, “Over $ 2 Billion Worth of Incentives Brought Intel to Ohio,” Axios Columbus, January 31, 2022, https://www.axios.com/local/columbus/2022/01/31/over-2b-worth-of-incentives-brought-intel-to-ohio. 141 Columbus State Community College, “Columbus State Leading New Ohio-Wide Community College Collaboration Creating Two-Year Degree Pathways to Chip Manufacturing Technician Careers at Intel,” September 8, 2022, https://www.cscc.edu/about/news/2022/intel-groundbreaking-sept-22.shtml. 142 Chris Bournea, “Intel Announces $50 Million Investment in Ohio Higher Education,” Ohio State University College of Engineering, March 18, 2022, https://engineering.osu.edu/news/2022/03/ intel-announces-50-million-investment-ohio-higher-education; Gabriela Cruz Thompson, “Intel Addresses Semiconductor Workforce Shortage,” September 24, 2023, https://newsroom. intel.com/corporate/intel-addresses-semiconductor-workforce-shortage. 143 Office of the Texas Governor, “Governor Abbott Announces Texas Semiconductor Innovation Fund Grant to Samsung Austin Semiconductor,” press release, September 17, 2025, https:// gov.texas.gov/news/post/governor-abbott-announces-texas-semiconductor-innovationfund-grant-to-samsung-austin-semiconductor. 81 End Notes 144 Vannevar Bush, “As We May Think,” The Atlantic, July 1945, https://www.theatlantic.com/magazine/archive/1945/07/as-we-may-think/303881. 145 Bush, “As We May Think.” 146 Rolfe Winkler et al., “Big Tech’s $ 400 Billion AI Spending Spree Just Got Wall Street’s Blessing,” Wall Street Journal, July 31, 2025, https://www.wsj.com/tech/ai/tech-ai-spending-companyvaluations-7b92104b. 147 John Jumper et al., “Highly Accurate Protein Structure Prediction with AlphaFold,” Nature 596 (2021): 583–589, https://doi.org/10.1038/s41586-021-03819-2. 148 Joseph L. Watson et al., “De Novo Design of Protein Structure and Function with RFdiffusion,” Nature 620 (2023): 1089–1100, https://doi.org/10.1038/s41586-023-06415-8. 149 Tong Wang et al., “Ab Initio Characterization of Protein Molecular Dynamics with AI2BMD,” Nature 635 (2024): 1019–1027, https://doi.org/10.1038/s41586-024-08127-z. 150 Jonas Degrave et al., “Magnetic Control of Tokamak Plasmas Through Deep Reinforcement Learning,” Nature 602 (2022): 414–419, https://doi.org/10.1038/s41586-021-04301-9. 151 ATLAS Collaboration, “Search for New Phenomena in Two-Body Invariant Mass Distributions Using Unsupervised Machine Learning for Anomaly Detection at √s=13 TeV with the ATLAS Detector,” Physical Review Letters 132, no. 8 (2024): 081801, https://doi.org/10.1103/Phys RevLett.132.081801. 152 Hiroaki Kitano, “Nobel Turing Challenge: Creating the Engine for Scientific Discovery,” npj Systems Biology and Applications, vol. 7, no. 29 (2021), https://doi.org/10.1038/s41540-021- 00189-3. 153 National Center for Science and Engineering Statistics, “Publication Output by Region, Country, or Economy and by Scientific Field,” in Publications Output: U.S. Trends and International Comparisons, NSB-2023-33 (National Science Foundation, December 11, 2023), https://ncses.nsf.gov/ pubs/nsb202333/publication-output-by-region-country-or-economy-and-by-scientific-field. 154 Conor Griffin et al., “A New Golden Age of Discovery: Seizing the AI for Science Opportunity,” AI Policy Perspectives (blog), November 26, 2024, https://www.aipolicyperspectives.com/p/ a-new-golden-age-of-discovery. 155 Haotian Teng et al., “Chiron: Translating Nanopore Raw Signal Directly into Nucleotide Sequence Using Deep Learning,” GigaScience 7, no. 5 (2018), https://doi.org/10.1093/gigascience/ giy037. 156 Ashish Vaswani et al., “Attention Is All You Need,” in Advances in Neural Information Processing Systems 30 (2017), 5998–6008. 157 Sayash Kapoor and Arvind Narayanan, “Could AI Slow Science? Confronting the ProductionProgress Paradox,” AI as Normal Technology (blog), July 16, 2025, https://www.normaltech.ai/p/ could-ai-slow-science. 158 “About,” Arc Institute, https://arcinstitute.org/about. 159 “2026 AI-for-Science Independent Postdoctoral Fellowship Program,” FutureHouse, https:// www.futurehouse.org/fellowship. 160 Anthropic, “Lawrence Livermore National Laboratory Expands Claude for Enterprise Use to Empower Scientists and Researchers,” July 9, 2025, https://www.anthropic.com/news/lawrence-livermore-national-laboratory-expands-claude-for-enterprise-to-empower-scientists-and; U.S. Department of Energy, National Nuclear Security Administration, “NNSA’s Los Alamos National Laboratory Launches Frontier AI Models on the Venado Supercomputer,” August 28, 2025, https://www.energy.gov/nnsa/articles/nnsas-los-alamos-national-laboratory-launches-frontier-ai-models-venado-supercomputer; National Aeronautics and Space Administration, Jet Propulsion Laboratory, “NASA’s Perseverance Rover Completes First 82 End Notes AI-Planned Drive on Mars,” January 30, 2026, https://www.jpl.nasa.gov/news/nasas-perseverance-rover-completes-first-ai-planned-drive-on-mars. 161 Exec. Order No. 14363, “Launching the Genesis Mission,” Federal Register 90 (November 28, 2025): 55035, https://www.federalregister.gov/documents/2025/11/28/2025-21665/launching-the-genesis-mission. 162 U.S. Department of Energy, “Energy Department Launches ‘Genesis Mission’ to Transform American Science and Innovation Through the AI Computing Revolution,” November 24, 2025, https://www.energy.gov/articles/energy-department-launches-genesis-mission-transform-american-science-and-innovation. 163 U.S. Department of Energy, “Energy Department Announces Collaboration Agreements with 24 Organizations to Advance the Genesis Mission,” December 18, 2025, https://www.energy.gov/ articles/energy-department-announces-collaboration-agreements-24-organizations-advancegenesis. 164 U.S. Department of Energy, Office of Science, The Transformational AI Models Consortium, DOE National Laboratory Program Announcement LAB 25-3560, August 22, 2025, https://science. osti.gov/-/media/grants/pdf/lab-announcements/2025/LAB-25-3560-000001.pdf. 165 The Materials Project, Lawrence Berkeley National Lab, https://next-gen.materialsproject.org. 166 Conor Griffin et al., “A New Golden Age of Discovery: Seizing the AI for Science Opportunity.” 167 Subcommittee on the Materials Genome Initiative, Materials Genome Initiative: Strategic Plan (National Science and Technology Council, Committee on Technology, December 2014), https:// www.mgi.gov/sites/mgi/files/mgi_strategic_plan_-_dec_2014.pdf. 168 Charles Yang, “Antitrust & the Science Instrument Industry,” The Republic of Science (blog), December 8, 2025, https://republicofscience.substack.com/p/antitrust-and-the-scienceinstrument. 169 U.S. Department of Energy, “Energy Department Advances Investments in AI for Science,” December 10, 2025, https://www.energy.gov/articles/energy-department-advances-investments-ai-science. 170 National Science Foundation, “Test Bed: Toward a Network of Programmable Cloud Laboratories (PCL Test Bed),” Solicitation NSF 25-541, July 16, 2025, https://www.nsf.gov/funding/ opportunities/pcl-test-bed-test-bed-toward-network-programmable-cloud-laboratories/nsf25- 541/solicitation. 171 Open Science Collaboration, “Estimating the Reproducibility of Psychological Science,” Science 349, no. 6251 (August 28, 2015): aac4716, https://doi.org/10.1126/science.aac4716. 172 Colin F. Camerer et al., “Evaluating Replicability of Laboratory Experiments in Economics,” Science 351, no. 6280 (March 3, 2016): 1433–1436, https://doi.org/10.1126/science.aaf0918. 173 Leonard P. Freedman et al., “The Economics of Reproducibility in Preclinical Research,” PLoS Biology 13, no. 6 (2015): e1002165, https://doi.org/10.1371/journal.pbio.1002165. 174 Exec. Order No. 14303, “Restoring Gold Standard Science,” Federal Register 90 (May 29, 2025): 22601, https://www.federalregister.gov/documents/2025/05/29/2025-09802/restoring-goldstandard-science. 175 “There was a 59.8% increase in ICLR submissions in 2025 alone” as stated in Jaeho Kim et al., “Position: The AI Conference Peer Review Crisis Demands Author Feedback and Reviewer Rewards,” Proceedings of the 42nd International Conference on Machine Learning 267 (2025): 81634–81651, https://proceedings.mlr.press/v267/kim25am.html. 176 Abel Brodeur and Bruno Barbarioli, “The Replication Engine,” Institute for Progress, The Launch Sequence, August 11, 2025, https://ifp.org/the-replication-engine. 83 End Notes 177 National Institutes of Health, “Replication to Enhance Research Impact Initiative,” NIH Common Fund, last reviewed February 4, 2026, https://commonfund.nih.gov/replication-initiative. 178 Mathlib community, “Completion of the Liquid Tensor Experiment,” Lean community blog, July 15, 2022, https://leanprover-community.github.io/blog/posts/lte-final. 179 Leila Sloman, “‘A-Team’ of Math Proves a Critical Link Between Addition and Sets,” Quanta Magazine, December 6, 2023, https://www.quantamagazine.org/a-team-of-math-proves-a-critical-link-between-addition-and-sets-20231206. 180 Math, Inc., “The Strong Prime Number Theorem,” GitHub repository, 2025, https://github.com/ math-inc/strongpnt. 181 Terence Tao, “A Conversation with Terry Tao, Inaugural Veritas Fellow,” interview by Jesse Han and Jared Duker Lichtman, Math, Inc., YouTube video, December 2025, https://www.youtube. com/watch?v=4ykbHwZQ8iU. 182 Seemay Chou, “Scientific Publishing: Enough is Enough,” Astera Institute (blog), June 2, 2025, https://asterainstitute.substack.com/p/scientific-publishing-enough-is-enough. 183 Timothy K. Mackey et al., “A Framework Proposal for Blockchain-Based Scientific Publishing Using Shared Governance,” Frontiers in Blockchain 2, (2019): 19, https://doi.org/10.3389/fbloc. 2019.00019. 184 “The Community of the DAO,” Nature Biotechnology 41 (2023): 1357, https://doi.org/10.1038/ s41587-023-02005-1. 185 Gaia Dempsey, “Why I Reject the Comparison of Metaculus to Prediction Markets,” Metaculus, February 24, 2023, https://www.metaculus.com/notebooks/17599/why-i-reject-the-comparison-of-metaculus-to-prediction-markets. Annex 85 American leadership in science and technology (S&T) underpins our economic prosperity, national security, and public health. As the United States celebrates the 250th anniversary of declaring its independence, we stand at the threshold of a new golden age of American innovation. The future of American leadership in the emerging technologies that will define this century, from frontier artificial intelligence (AI) to quantum technologies and advanced nuclear fission and fusion, depends in part on core Federal investments in foundational research, the basic and use-inspired inquiry upon which a broad range of sciences and engineering work depends. Rapid technological advances are transforming the way scientific research is conducted, the scientific questions that we can now ask, and the scientific instruments we can build. To usher in this new golden age, we must renew the research and development (R&D) enterprise on which our scientific leadership depends. Eighty years ago, Vannevar Bush’s Science: The Endless Frontier laid the foundation for the modern American scientific enterprise, giving rise to the National Science Foundation and a partnership between Federal Government, universities, and industry that won the American Century. Today, that enterprise is being MEMORANDUM FOR THE HEADS OF EXECUTIVE DEPARTMENTS AND AGENCIES NSTM-5 / M-26-16 July 21, 2026 FROM: SUBJECT: MICHAEL J. KRATSIOS ASSISTANT TO THE PRESIDENT FOR SCIENCE AND TECHNOLOGY DIRECTOR, OFFICE OF SCIENCE AND TECHNOLOGY POLICY RUSSELL T. VOUGHT DIRECTOR, OFFICE OF MANAGEMENT AND BUDGET Ushering in a New Golden Age of American Innovation: Fiscal Year 2028 Administration Research and Development Budget Priorities 86 Annex: FY 2028 R&D Priorities Memo reshaped by forces Bush could not have foreseen. Global competitors are racing to challenge U.S. scientific leadership, developing new methods to drive discovery and innovation. At the same time, our own enterprise has fallen out of balance. Industry now drives a growing share of innovation and even basic research, where its share of national R&D funding has doubled over the past half century,and yet we have largely not updated how the Federal Government funds research or partners with the private sector. The government invests more in R&D than ever before, yet much of the non-defense increase is concentrated in the life sciences and the pace of significant breakthroughs has slowed. And while transformative discoveries are still made in America, too often we fail to capitalize on them at home, ceding the manufacturing and supply chains that turn discovery into industry to competitors abroad. The opportunity before us is clear: by integrating industry more fully into the research enterprise, funding transformative science, especially in the physical sciences and engineering, and reconnecting scientific discovery with manufacturing and skilled crafts, America can once again fully translate scientific discovery into broad-based prosperity, creating new applications, high-paying jobs, and stronger regional economies. This memorandum provides guidance to Federal departments and agencies (agencies) to recalibrate the Nation’s S&T enterprise, implementing the recommendations in Science: A New Golden Age and advancing the President’s vision of a Golden Age of American Innovation. The guidance identifies Administration R&D priorities for agencies to consider, as appropriate, in Fiscal Year (FY) 2028 Budget formulation and related planning. These priorities include: (i) rebalancing R&D portfolios toward foundational research and the physical sciences and engineering, (ii) advancing national S&T missions, (iii) applying AI and emerging technologies to accelerate American research and innovation, (iv) expanding R&D infrastructure for broader ecosystem use, (v) translating scientific advances into stronger regional ecosystems and broad-based prosperity, (vi) considering new funding mechanisms and institutional models to support frontier science, (vii) exploring better ways to identify and develop scientific talent, (viii) rigorously studying, evaluating, and improving how Federal science is funded, and (ix) integrating Federal R&D into the broader S&T enterprise. Agencies should account for this guidance, as appropriate, in their FY 2028 Budget submission to the Office of Management and Budget (OMB). 87 Annex: FY 2028 R&D Priorities Memo FY 2028 R&D PRIORITY AREAS Invest in Foundational Research to Drive Scientific Breakthroughs for Emerging Technologies Foundational research, including basic and use-inspired inquiry across the sciences and engineering, remains the bedrock of American scientific and technological leadership. The United States derives outsized long-term security, economic, and societal returns from foundational research, which expands the frontier of knowledge and leads to the growth of new industries. The Federal Government’s comparative advantage relative to private industry lies here, in supporting work where payoffs are long-horizon, broadly distributed, and difficult to realize privately. In their FY 2028 budget submissions to OMB, agencies should seek to increase the share of foundational research relative to later-stage development. Many of the Administration’s strategic technology priorities, including AI, quantum information science, semiconductors, advanced communications, robotics, advanced manufacturing, nuclear fission and fusion, and space systems, all rely on foundational research across the physical sciences, computer science, and engineering. However, the physical sciences and engineering have declined as a share of the Federal research portfolio over an extended period, even as the strategic importance of these fields has grown. Agencies are encouraged to prioritize both the absolute level and the relative share of funding directed within budget guidance levels to the physical sciences (physics, chemistry, materials science, space science, etc.), computer science, and supporting engineering and mathematical disciplines, especially within national security-relevant research portfolios. In addition, to support Administration priorities in biotechnology and biomanufacturing, agencies should prioritize foundational research in the biological sciences over the life sciences, a broader category not focused on foundational research. In their FY 2028 Budget submissions to OMB, agencies should note the R&D character classification of proposed activities as a percentage of their R&D funding portfolio and identify the specific programs through which the agency proposes to shift its portfolio toward earlier-stage work. Where agencies propose to significantly expand later-stage development activities, they should justify why such activities would not occur absent Federal support. Agencies should prioritize funding for: • Physical Sciences. Agencies should prioritize foundational research in the physical sciences, including condensed matter and quantum materials 88 Annex: FY 2028 R&D Priorities Memo physics, including correlated, magnetic, and topological states; photonics, addressing the generation, control, and detection of light; atomic, molecular, and optical physics, addressing precision measurement and the quantum control of systems; the physics of superconductivity and other quantum phenomena; plasma and high energy density physics; nuclear physics and matter under extreme conditions; gravitational physics and geodesy; and space and planetary physics, including the radiation, plasma, and space-weather conditions in which space systems operate. These fields underpin quantum science, semiconductors, advanced communications networks, future computing technologies, advanced nuclear fission and fusion energy, and space exploration technologies including novel sensing modalities and precision position, navigation, and timing. • Chemistry and Materials Science. Agencies should prioritize foundational research in chemistry and materials science, including electronic, photonic, and quantum materials; the surface, interface, and defect chemistry that governs fabrication and device performance; materials for extreme environments (e.g., radiation-tolerant, plasma-facing, and high-temperature); the structure, properties, synthesis, and characterization of materials, including condensed matter and materials theory, ceramics, metals, polymers and biomaterials; electrochemistry and solid-state ionics; and catalysis, synthesis, and reaction mechanisms. These fields underpin quantum science and semiconductors and extend across advanced manufacturing, energy production and storage, the nuclear fuel cycle, photonics, and space and hypersonic systems. • Mathematics and Computer Science. Agencies should prioritize foundational research in the mathematical and computational sciences, including applied and computational mathematics, numerical analysis and uncertainty quantification; classical and quantum information theory; algorithms, computational complexity, and cryptography, including post-quantum cryptography; the mathematics of optimization and control; statistics, probability, and the foundations of data science; and the foundations of high-performance and future computing. These fields underpin advanced communications networks and secure information systems, quantum information science and future computing, and the modeling, simulation, and verification on which fusion energy, advanced manufacturing, and space systems depend. 89 Annex: FY 2028 R&D Priorities Memo • Engineering Sciences. Agencies should prioritize foundational research in engineering sciences, including microelectronics, photonic, quantum, and microsystem device engineering and early-stage manufacturing; the electromagnetic, radiofrequency, and propulsion sciences; the thermal, fluid, and mechanical sciences, including solid mechanics and the mechanics of materials; the dynamics, estimation, and control of complex systems, including astrodynamics, guidance, and navigation; and magnet, superconducting, and power-system engineering. These fields underpin semiconductors and advanced communications networks, advanced manufacturing, space systems, robotics, and fission and fusion energy. • Biological Sciences. Agencies with general, broad-based life-sciences research missions should prioritize foundational research in the biological sciences including molecular, cellular and structural biology; biochemistry and chemical biology; genetics, genomics, and synthetic and engineering biology; neuroscience and the neural basis of cognition and behavior; and microbiology and quantitative biology. These fields underpin biotechnology and biomanufacturing, neurosciences and brain-machine interfaces, and human health and therapeutics. Advance National Science and Technology Missions From the Manhattan Project to the Apollo Program, some of America’s greatest scientific achievements have come from focused national missions that united the Nation’s brightest minds behind an ambitious common goal. This Administration has revived that mission-driven model for a new era of global competition, launching a set of national science and technology efforts targeting the technologies that will define the coming century. Federal R&D is uniquely suited to drive these efforts forward by supporting them across every stage from foundational discovery to demonstration, sustaining the long-horizon and high-risk work the private sector cannot undertake alone, and convening the partnerships among government, industry, academia, and philanthropy through which national missions are ultimately achieved. Realizing them will demand a comparable concentration of national effort. Agencies should align their R&D investments, where appropriate, with the Administration’s national missions, including: • AI: The Genesis Mission to harness AI to double the productivity and impact of America’s research enterprise within a decade, including agencyspecific contributions across national S&T challenges and compute and 90 Annex: FY 2028 R&D Priorities Memo research infrastructure for the American Science and Security Platform, pursuant to Executive Order 14363; • Quantum: The Quantum Computer for Application Development and Discovery Science (QC-ADDS) effort to develop a quantum computer at a scale intended to initiate the era of quantum-enabled scientific discovery, pursuant to Executive Order 14413; • Fusion: Demonstration of commercial fusion power in the United States by the mid-2030s, following the Department of Energy’s Fusion Science & Technology Roadmap; • Space: Return of Americans to the lunar surface by 2028, the construction of a lunar base, the National Initiative for American Space Nuclear Power, and the development of a responsive and adaptive national security space architecture, pursuant to Executive Order 14369; • Robotics: General-purpose autonomous systems capable of dexterous manipulation, mobility, and reliable operation in real-world environments, to initiate the era of physical AI-driven scientific discovery and American reindustrialization; and • Semiconductors: Next-generation semiconductor technologies, including EUV-and-beyond photolithography, 3D advanced packaging, and novel materials for future semiconductor devices and technology nodes. Agencies should support these missions through the full range of R&D policy instruments available to them. Each agency should identify, through the FY 2028 Budget process and other established budget review channels how its missionspecific research priorities and programs can support these national goals, consistent with statutory authorities, agency missions, and available resources. In their FY 2028 budget submissions, agencies should consider how to prioritize their R&D infrastructure, including user facilities, testbeds, and high-performance computing assets, toward mission needs and expand access for university and industry partners. Agencies should also propose investments that employ the full set of talent and incentive mechanisms at their disposal, including graduate and postdoctoral fellowships to build the skilled workforce these missions require, and prizes, grand challenges, and competitions to mobilize the broadest possible range of innovators toward the hardest problems. 91 Annex: FY 2028 R&D Priorities Memo Build the Foundation for a New Era of Scientific Discovery AI and emerging technologies have immense potential to transform science by unlocking novel experimental and analytical capabilities, enabling new ways to organize the research enterprise, and prompting new fields of scientific inquiry. In November 2025, President Trump launched the Genesis Mission, a whole-ofgovernment effort to harness the AI-driven computing revolution with the intent to double the productivity and impact of American science and engineering within a decade. Rather than crowding subfields of AI research where private capital is already abundant, the Genesis Mission is designed to ensure America’s scientific enterprise is first and fastest to harness these technologies for discovery across the scientific landscape. Agencies should identify opportunities to integrate AI and other emerging technologies into research as appropriate; prepare Federal scientific instrumentation, datasets, and compute for the AI-for-science transformation; and treat support for the Genesis Mission as a central R&D priority. Proposed agency efforts in this area should be noted in FY 2028 Budget submissions. Agencies should prioritize funding for: • AI as an Instrument of Scientific Discovery. Agencies should fund research that uses AI as a new instrument of scientific discovery, not merely as a tool to augment existing capabilities. Agencies should seek out proposals that thoughtfully integrate AI into scientific workflows, rather than projects that apply AI for incremental gains or without clear justification for why the problem requires AI-specific methods. Agencies should align R&D funding with the Genesis Mission’s National S&T Challenge areas where appropriate and propose new or expanded challenges consistent with their own priorities. Given the scale of private sector investment in AI, agencies should prioritize work that industry is unlikely to pursue on its own, including pre-competitive research outputs and enabling platform technologies. • Scientific Foundation Model Development. Agencies should propose investments that support domain-specific scientific foundation models that enable high-fidelity simulations of natural phenomena and accelerate scientific discovery across Genesis Mission’s National S&T Challenge areas, including advanced manufacturing, biotechnology, critical materials, nuclear fission and fusion, quantum information science, and semiconductors, and coordinate with other agencies as applicable. These models require curated scientific datasets and compute that no performer 92 Annex: FY 2028 R&D Priorities Memo can assemble alone, making the Federal Government uniquely wellpositioned to develop them as shared, pre-competitive assets for the research community. • Scientific Data Generation for AI. Agencies should propose efforts to make internal scientific datasets available for use and investments in the data infrastructure that makes them accessible for AI training and inference. Agencies should create incentives for researchers to curate and share valuable data that is routinely abandoned due to lack of dedicated funding or recognition, including experimental records, negative results, and operational data from laboratory procedures. Agencies should further support the creation, curation, and stewardship of ambitious new datasets that could open entirely new fields of inquiry or deliver exceptional value to the Nation’s S&T enterprise. As laboratory automation matures, agencies should propose investments in infrastructure to capture data at an industrial scale, laying the groundwork for a future of rapid, autonomous scientific discovery. • Integration of AI with Scientific Instrumentation. Agencies should build on the Genesis Mission by proposing investments in robotics, automated laboratories, modernization of user facilities to operate within closed-loop AI scientific workflows, and autonomous control of large-scale experiments in which AI systems generate hypotheses, conduct experiments, interpret results, and iterate in real time. Agencies should leverage their purchasing power to build domestic supply chains for AI-ready scientific instrumentation and drive the redesign of these instruments with open interfaces, standardized data formats, and cross-vendor interoperability, making it easier for researchers to connect instruments and use the software tools best suited to their work. Expand World-Class R&D Infrastructure for Broad Use The productivity of Federal R&D depends on scientific infrastructure, including the physical platforms, user facilities, instrumentation, compute, and laboratory spaces through which research is conducted. These assets have long planning horizons, high fixed costs, and operating requirements that extend well beyond the grants they support, and are often out of reach for individual investigators and institutions. When broadly accessible, this infrastructure enables scientists to pursue cutting-edge research and focus on conducting their best science, rather than the time and capital required to build their own infrastructure and facilities. 93 Annex: FY 2028 R&D Priorities Memo In their FY 2028 budget submissions to OMB, agencies should assess scientific infrastructure needs deliberately rather than treating them as a residual claim on research grants. In particular, agencies should propose investments in mid-scale instrumentation, fully funded within a fiscal year and aligned with Administration priorities, given it has historically been underfunded relative to its scientific importance; advanced compute; and sustained operating support for user facilities and shared platforms. Where agencies propose to significantly reduce or defer these investments, they should justify the proposal and explain how they will address the resulting gaps and sustain operation of existing facilities. To expand the reach of investments in scientific infrastructure, agencies should prioritize funding for: • User Facilities for the S&T Ecosystem. Agencies should propose investments in cutting-edge R&D infrastructure and instrumentation to enable researchers and innovators to validate new hypotheses, test prototypes, and scale new technologies, lowering barriers to frontier research. Proposed investments should be consistent with overarching Administration priorities to both maximize the use of existing infrastructure by addressing deferred maintenance and increase efficiency by reducing footprints and when necessary, include new infrastructure to achieve the greatest utilization by a broad community of researchers, including the private sector and other non-Federal researchers. Agencies should consider the resources needed to increase access to Federal R&D facilities by adopting evaluation criteria that weigh innovative potential and commercial urgency alongside scientific merit, streamlining Cooperative Research and Development Agreements and licensing processes, and reducing administrative burdens on industry users. These arrangements should encourage facilities to leverage industry cost-share arrangements and user-fee revenue to expand capacity and fund next-generation instrumentation. • Advanced Compute for Federal R&D. Compute is the foundation of AIenabled science, and Federal infrastructure must keep pace with the scale and flexible access researchers now require. Agencies should propose investments that expand access to advanced compute infrastructure, including unified access portals, standardized applications, and common data and software environments that allow researchers to move work seamlessly across facilities. Application processes should lower the barrier to entry for students, individual investigators, and small teams, particularly for fast-turnaround projects. Federal compute investment should offer 94 Annex: FY 2028 R&D Priorities Memo capabilities differentiated from the commercial market, such as highly secure data centers for sensitive research, access to unique Federal datasets, and specialized AI accelerators and computing architectures. Where commercial compute is cost-effective and meets researcher needs, agencies should pursue public-private partnerships or procure capacity through commercial providers to improve agility and time-to-science. Leverage R&D to Strengthen Regional Manufacturing and Industry Federal R&D investments can be leveraged to translate scientific discoveries into benefits for all Americans, securing broad-based prosperity and supporting the reindustrialization of our Nation. Achieving these objectives require Federal investments that pair foundational research with advanced manufacturing, strengthen regional ecosystems, build resilient supply chains, develop a skilled technical workforce, and catalyze non-Federal investment to the greatest extent possible. Agencies should prioritize funding for manufacturing R&D across strategic technologies with the goal of building domestic manufacturing capacity and supply chains to produce the next generation of semiconductors, advanced materials, biotechnology, nuclear technologies, and robotics. Manufacturing R&D spans the full research spectrum: the manufacturing science underlying how things are made, including process science, materials science, metrology, automation, and the underlying physics, chemistry, and engineering; advanced engineering methods and production technologies; translational programs such as manufacturing innovation institutes, pilot lines, and demonstration facilities that bridge laboratory discovery and production; and supply chain analytics. Cost-share arrangements should generally be considered, and where appropriate, expected for later-stage manufacturing and demonstration activities, while earlier-stage manufacturing science should be supported on terms appropriate to foundational research. 95 Annex: FY 2028 R&D Priorities Memo R&D PRIORITY PRACTICES

1. Develop New Mechanisms to Support Frontier Science America must continue to expand the repertoire of institutions and R&D funding mechanisms it uses to conduct science, enabling our best researchers to tackle the most ambitious S&T challenges that exist. These mechanisms should account for forces reshaping the scientific enterprise, including the rise of funding from industry and philanthropy, and the growing importance of genuinely integrated, multidisciplinary teams. The Federal Government should incentivize new institutional models that complement conventional principal-investigator driven laboratories, industry laboratories, and Federal R&D facilities. It should also supplement conventional, consensus-driven peer review, which excels at advancing established lines of inquiry, with new review mechanisms that are better suited to recognizing high-risk, high-reward research, early-career talent, or ideas that fall outside established disciplinary boundaries. Agencies should adopt a deliberate, portfoliobased approach that matches funding mechanisms to the S&T challenges they seek to address, maximizing Federal return on investment through an explicit mix of modalities, risk profiles, and time horizons. A broader menu of institutional structures and funding mechanisms will enable new forms of scientific work, encourage scientists to pursue novel lines of inquiry, and attract highercaliber reviewers empowered to make bold bets. Agencies are encouraged to review their existing institutional models and funding mechanisms, explore new ones to close gaps in areas critical for national priorities, and construct balanced Federal R&D portfolios according to the following principles: • Support a Diverse Portfolio of Institutions. The Federal Government should reflect a portfolio of institutions that collectively advance the core objectives of the Nation’s S&T enterprise, including conducting a range of scientific work, training the next generation of scientists, and translating scientific discoveries into concrete benefits for Americans. Agencies should identify objectives that remain unaddressed because no existing institution is well-suited to pursue them. One notable gap is agile, mid-scale science: infrastructure-heavy, multidisciplinary basic research that requires coordinated teams of ten to a hundred people. Agencies have begun to address these gaps through new models like the U.S. National Science Foundation’s (NSF) X-Labs and certain Advanced Research Projects Agency 96 Annex: FY 2028 R&D Priorities Memo programs. Agencies should consider these models and experiment with additional designs to enable new types of scientific pursuits. • Increase Grant Durations for Transformative Research. Agencies should develop proposals for the FY 2028 Budget that would expand the number of long-duration grants, ideally lasting five years or more, that give our best researchers the time and autonomy to pursue bold, ambitious projects whose most important results may take years to emerge. These awards should be fully-funded in year one, with all resources earmarked upfront, to minimize administrative burdens and reduce pressure for researchers to generate intermediate results to secure continued funding. This upfront commitment should be paired with clear performance metrics and periodic reviews, with the understanding that funding may be withdrawn and redirected if needed. Existing programs, such as the National Institute of Health (NIH) Director’s Pioneer Award and the DOW Vannevar Bush Faculty Fellowship, offer useful models for long-duration, investigator-centered support for creative basic research. • Expand Use of Fast Grants for Exploratory Projects. Agencies should consider establishing or expanding, where authorized and consistent with available resources, flexible, low-friction “fast grants” to support preliminary research, exploratory projects, and time-sensitive work. These programs should feature simplified applications requiring just a few pages of writing, rapid review timelines of under one month, and award sizes calibrated to proof-of-concept work. Agencies should encourage greater use of existing mechanisms and ensure that they meet their intended timelines, while developing additional fast-track pathways as needed. • Design Ambitious Prizes and Challenges. Well-designed prizes can spur cross-disciplinary collaboration, attract nontraditional entrants, mobilize substantial private capital, and catalyze entirely new industries with a relatively small amount of funding. Agencies should expand their use of prizes and challenges to advance national missions. To maximize participation from nontraditional teams, agencies should emphasize outcomebased goals rather than prescribing specific methods. Prizes should target at least a 3:1 leverage of private to Federal investment, and may be paired with complementary incentives such as advance procurement commitments, regulatory fast-tracking, and access to Federal testing facilities. 97 Annex: FY 2028 R&D Priorities Memo • Experiment with Emerging Funding Mechanisms. Agencies should study, pilot, and evaluate whether there are existing models or additional designs for innovative funding mechanisms beyond those described above and in conjunction with OMB and OSTP. Mechanisms for consideration could include, as appropriate, “golden tickets” that let individual agency technical reviewers recommend unconventional proposals that may not pass consensus-driven review panels, which tend to skew toward funding more incremental advances; advance market commitments that signal demand for a scientific or technical capability before it exists, subject to available appropriations and demonstration of capabilities against clearly defined criteria; regranting models that delegate funding authority to working scientists to tap distributed expertise; and more speculative approaches such as quadratic funding or eigenfunding. Such models could surface valuable ideas too divisive for committees and attract higherquality reviewers by empowering them to exercise independent scientific judgment. Such approaches will not be a one-size-fits-all solution to grantmaking, but should be appropriately explored for their potential role in the Federal R&D portfolio as agencies look to more effectively support the American S&T enterprise. Agencies should ensure that new funding mechanisms strictly adhere to agencies’ legal authorities and conflict of interest policies, and that funded proposals meet a level of scientific rigor appropriate for Gold Standard Science.
2. Identify and Develop Top Technical Talent The Federal Government should orient around the scientists, engineers, and technicians who serve our Nation, providing them with the support, freedom, and opportunities needed to do their best work. S&T workforce programs should select the best and brightest Americans, recognizing that these individuals are distributed across the Nation, not isolated to major metropolitan areas. The programs should identify and invest early in high-potential students and early-career individuals, while cultivating their long-term commitment to America’s S&T enterprise. Agencies with S&T workforce development programs, including graduate fellowships; K-12 Science, Technology, Engineering, and Mathematics (STEM) education; and skilled technical workforce programs should review existing efforts and, where appropriate, propose modifications or new approaches through established budget and policy processes to: 98 Annex: FY 2028 R&D Priorities Memo • Identify Exceptional Talent Nationwide. S&T workforce programs should identify and support all talented Americans across geographies, incomes, and demographics. Exceptional talent is defined by demonstrated technical ability, not background or identity. Agencies should therefore anchor selection processes in criteria predictive of STEM success, such as reasoning assessments, domain competitions, engineering portfolios, and technical work, rather than relying on self-selection, essays, institutional referrals, or polish and credentials. Agencies should leverage merit-based identification mechanisms that cover as many people as possible (e.g., SAT scores or other standardized quantitative assessments) to find overlooked talent. • Expand Advanced K-12 STEM Enrichment Opportunities. Targeted programs can accelerate the development of advanced K-12 STEM talent by increasing exposure to pathways into scientific careers and connecting students with expert mentors and similarly capable peers. Where appropriate, agencies should support K-12 STEM enrichment opportunities, such as residential math and science programs and Olympiad-style competitions, that immerse high-ability students in advanced S&T environments and direct their ambitions to the hardest open questions. • Expand Hands-On Technical Learning. S&T workforce programs should provide early and sustained exposure to real-world technical environments. Agencies should treat research placements in academic, industry, and Federal laboratories as standard components of high-quality S&T talent development programs. Placements should be substantive, last at least one semester, and provide participants with meaningful access to advanced scientific instrumentation, datasets, and challenges not available in traditional academic settings. Agencies should expand opportunities for hands-on technical learning as early as high school through work-based learning, vocational training, makerspace access, and machine shop classes. • Support Early-Career Researchers. Agencies should strengthen support for graduate students, post-doctoral researchers, and early-career faculty, when research creativity is often highest but institutional support the weakest. Agencies should expand the use of fellowship programs and address conditions that limit the mobility of graduate students, post-doctoral researchers, and early-career researchers as they navigate opportunities in the S&T enterprise. These programs can provide young scientists with 99 Annex: FY 2028 R&D Priorities Memo resources and intellectual freedom during the most pivotal stage of their careers, encouraging them to remain in the Nation’s S&T enterprise. • Support Individuals Agnostic of Institutional Affiliations. S&T workforce programs should provide individuals flexibility to choose their research institutions, supervisors, and topics. Agencies should prioritize programs that distribute funding directly to students and researchers, similar to NSF’s Graduate Research Fellowship Program, so recipients can apply the grant to any qualifying institution that best supports their goals and retain it if they move, encouraging institutions to compete for earlycareer talent. Agencies should develop the capability to track supported individuals longitudinally across multi-year transitions, minimizing the need for individuals to re-discover and re-apply for support. • Build Flexible Cross-Sector Talent Pathways. The Nation’s top scientific talent should be encouraged to gain experience across research cultures and engineering environments throughout their careers. Agencies should expand opportunities for scientists, engineers, and skilled technical workers to move fluidly across academia, industry, and Federal R&D facilities by increasing the flexibility of academic fellowships, establishing crossinstitution placements like joint industry or Federal laboratory Ph.D. programs, and supporting alternative paths for skilled technical workers to participate in academic training and scientific discovery. • Encourage Broad Post-Fellowship Service. Federal investments in individuals should strengthen the Nation’s S&T enterprise. Agencies should consider incorporating service requirements into fellowship programs while defining service broadly to capture the myriad ways individuals can leverage their training to advance that enterprise. Qualifying service could include academic research and training the next generation of American scientists, entrepreneurship, work in the defense industrial base, advisory roles that shape Federal S&T priorities, or government and military service. Agencies should aim to make any service requirements flexible enough for recipients to pursue the highest-impact opportunities after their fellowship ends and to attract the strongest candidates. 100 Annex: FY 2028 R&D Priorities Memo
3. Build a Self-Improving Scientific Enterprise The Federal Government invests approximately two hundred billion dollars in R&D each year, but allocates comparatively little to understanding which funding mechanisms, institutional models, workforce development programs, and research practices produce the strongest scientific outcomes. Agencies should treat the science of science-funding with the same rigor as the science they fund, and build the organizational capacity to learn, experiment, and improve continuously. Agencies should assess whether to establish metascience capabilities, where appropriate, following the guidance below: • Establish Metascience Capabilities. Agencies should establish metascience capabilities that evaluate what programs actually work and drive organization-wide reforms. Core responsibilities should include conducting research on how factors such as funding mechanisms, peer review, and publication practices affect scientific outcomes; piloting novel funding mechanisms and institutional models; and evaluating pilots and informing agency-wide portfolio management. These functions should be established at a sufficiently high level within agencies to effect real, crossagency change. • Develop Systematic Gap-Mapping Capacity. Agencies should develop the capacity to systematically compare their grantmaking portfolios against the landscape of unsolved scientific and technical challenges in their domains, rather than relying primarily on historical funding patterns. In collaboration with industry, academia, and philanthropy, agencies should maintain “gap maps” that identify unmet needs, duplicated efforts, and emerging opportunities. Gap maps should directly inform portfolio construction, helping agencies select appropriate funding mechanisms and institutional models to target the most important and neglected gaps. • Build Data Infrastructure for Metascience. Agencies should develop purpose-built data infrastructure for metascience, including systems that integrate application-level data, reviewer behavior and scoring, and links between awards and downstream outcomes. These systems should support longitudinal tracking for both awardees and near-miss applicants. Agencies should assess workforce and contracting operations for software engineers and data scientists with the skills to build and maintain these systems as a core institutional capability. 101 Annex: FY 2028 R&D Priorities Memo • Elevate and Empower Agency Program Officers. The effectiveness of Federal R&D funding depends heavily on agencies’ ability to recruit exceptional program officers and give them genuine discretion to define technical problems, build a research portfolio, and manage toward ambitious outcomes. Agencies should consider approaches for recruiting top scientists, engineers, entrepreneurs, and philanthropists into time-limited public service and raising their prestige, visibility, and authority. Agencies should also assess options for reducing barriers to hiring program officers from non-traditional backgrounds, expanding rotational mechanisms such as the Intergovernmental Personnel Act, and developing competitive compensation and career pathways that make program management a careerenhancing opportunity for top scientific talent. Agencies should also develop or enforce mechanisms to ensure that conflict of interest policies are strictly followed for all employees involved in funding recommendations and decisions. • Reduce Administrative Burdens. Agencies should reduce administrative burdens in the grantmaking and research process to maximize the impact of taxpayer-funded science. This includes clarifying requirements for the research community and eliminating overcompliance beyond what Federal regulations and statutes require. Agencies should consider proposals to coordinate to harmonize and standardize grant requirements, forms, and submission processes to the greatest extent possible, and carefully weigh any incremental gains in oversight from new requirements or regulations against the cumulative burden they impose on researchers. Agencies should also consider options for easing administrative and regulatory burdens on Federal technology transfer to increase private-sector investment in R&D.
4. Integrate Federal R&D into Broader S&T Enterprise Federal R&D is one part of a far larger national S&T enterprise that spans private industry, academia, state and local governments, and the regional economies in which discovery is translated into production. To maximize the return on Federal investment, agencies should more deliberately integrate their R&D with this broader enterprise. This means looking for opportunities to expand the use of non-Federal cost share, so that Federal dollars draw in and are amplified by private and other non-Federal investment rather than standing alone. It also means coordination between Federal R&D and non-R&D investments to support the 102 Annex: FY 2028 R&D Priorities Memo growth of regional innovation ecosystems and domestic manufacturing hubs consistent with statutory purposes. • Drive Greater Integration of Foundational and Applied Research. In many frontier technologies, scientific discovery, engineering, and manufacturing R&D are not sequential but iterative and tightly coupled. Where appropriate, agencies should propose funding consortia and partnerships that integrate basic research with manufacturing R&D, reflecting the multidisciplinary, engineering-intensive way science is conducted today. Agency funding in this area should ensure the pursuit of long-term research agendas in partnership with industry, employment of career scientists, engineers, and technicians, publication of foundational discoveries as public goods while licensing specific process innovations, and co-locate with manufacturing facilities and testbeds. Agencies should explore how such institutions can provide durable infrastructure to anchor place-based innovation ecosystems aligned with a region’s economic strength. • Expand the Use of Non-Federal Cost Share. Federal R&D funding is most effective when it catalyzes, rather than substitutes for, private and nonFederal investment. Agencies should structure funding opportunities, within existing resources, to prioritize support for initiatives that incorporate meaningful non-government cost share from industry, philanthropy, State and local governments, or international partners. Cost-share arrangements signal market validation, accelerate translation, distribute risk, and extend the impact of taxpayer-funded research. These arrangements should draw on the deep domain expertise external funders have built in particular sub-fields and leverage their networks to identify exceptional grant opportunities. Agencies should review existing authorities for costshared R&D, including cooperative agreements, public-private partnerships, consortia models, and other transaction authorities where applicable, and propose expansions where statutory or regulatory barriers can be addressed. • Integrate Federal R&D with Non-R&D Investments to Support Regional Ecosystems. The impact of Federal R&D depends critically on the surrounding ecosystem, including the workforce, infrastructure, capital, supply chains, and institutions that translate discovery into economic growth. Agencies should coordinate R&D investments with Federal nonR&D investments, including in workforce and education, economic 103 Annex: FY 2028 R&D Priorities Memo development, infrastructure, small business support, manufacturing extension, and procurement, to strengthen regional innovation ecosystems and ensure that the benefits of Federal science are broadly distributed across American communities, particularly where doing so would accelerate industry-specific R&D anchored in a region’s area of expertise. Agencies should coordinate across the Federal Government, including through OMB, the NSTC, and agency-to-agency agreements where helpful, to identify opportunities to co-locate, sequence, or jointly award R&D and non-R&D resources in support of place-based strategies and ensure complementary Federal investments in a given region. • Integrate Industry in Workforce Training. S&T workforce programs should maximize collaboration with the private sector, which increasingly leads both basic and applied R&D, holds unique scientific instrumentation, data, and computing resources, and can recruit the best science and engineering talent in ways no university can match. Where practicable, agencies should partner with industry to attract stronger applicants and amplify Federal investments, including through industry co-funding (e.g., tuition, stipends, and portable research funding), paid internship placements, access to research infrastructure, curriculum development, and expert mentorship. 104 Annex: FY 2028 R&D Priorities Memo IMPLEMENTATION To address the budget formulation priorities set forth in the “FY 2028 R&D Priority Areas” section of this memorandum, agencies should follow the standard process for FY2028 budget submission to OMB. In addition, within 90 days of this memorandum, the head of each agency with $3 billion or more in FY 2026 budget authority for R&D shall submit to the Assistant to the President for Science and Technology (APST) and Director of the Office of Management and Budget (OMB Director) an action plan describing how the agency intends to implement the program implementation guidance set forth in the “R&D Priority Practices” section of this memorandum. Agency action plans should identify how program execution of their FY2026 and FY2027 budgets can support these priority practices. Budget formulation matters addressed by this memorandum are outside the scope of action plans and should instead be reflected in agency FY 2028 budget submissions to OMB. Each action plan shall identify specific actions to address each R&D priority practice (e.g., new funding opportunities, program solicitations, pilot initiatives, statements to the research community, internal organizational changes), implementation timelines, and the offices responsible. OSTP and OMB will coordinate implementation of these action plans and issue supplementary guidance as appropriate.