Executive Summary
On July 21, 2026, the White House Office of Science and Technology Policy (OSTP) released a landmark report titled “Science: A New Golden Age,” authored by OSTP Director Michael Kratsios. The report represents the most ambitious re-examination of America’s scientific research and development (R&D) ecosystem since Vannevar Bush’s canonical 1945 report Science: The Endless Frontier, which established the National Science Foundation (NSF) and shaped U.S. science policy for eight decades.
The 125-page report delivers a sobering diagnosis of systemic dysfunctions in the American scientific enterprise — including slowing breakthrough rates, crippling administrative burdens, misaligned incentives, an eroded industrial base, and intensifying global competition — while also charting an aggressive path forward across five interconnected strategic pillars: revitalizing the research enterprise, securing technological dominance, broadening the benefits of science, preparing for the AI revolution, and fundamentally restructuring how the federal government funds and conducts science.
Accompanying the report is a formal FY 2028 R&D Budget Priorities Memorandum (NSTM-5 / M-26-16), co-signed by OMB Director Russell Vought, which translates the report’s recommendations into binding budget guidance for all federal departments and agencies.
This article provides a comprehensive, standalone analysis of the report’s complete content — including all key data points, strategic recommendations, institutional proposals, and budget priorities — designed for readers, policymakers, and researchers who need the full picture without access to the original document.
1. Historical Context: From The Endless Frontier to A New Golden Age
The Vannevar Bush Legacy
In 1945, as World War II drew to a close, President Franklin D. Roosevelt commissioned his science advisor Vannevar Bush to envision how America’s wartime scientific mobilization could be redirected toward peacetime prosperity. Bush’s response — Science: The Endless Frontier — made the case for sustained federal investment in basic research, leading directly to the creation of the National Science Foundation and the tripartite partnership of government, academia, and industry that powered the American Century.
Bush’s core insight was that the federal government possessed a unique comparative advantage in funding basic research — work where payoffs are long-horizon, broadly distributed, and difficult for private actors to capture. This “linear model” of technical progress (basic research → applied research → development → industry) became the organizing principle of the post-war scientific enterprise.
Why a New Report Now — 81 Years Later
President Trump’s letter of March 26, 2025 — delivered on the occasion of Kratsios’s Senate confirmation as OSTP Director — explicitly invoked the FDR-Bush precedent:
“Just as FDR tasked Vannevar Bush, I am tasking you with meeting the challenges below to deliver for the American people.”
The letter posed three foundational questions:
| Priority | Presidential Question |
|---|---|
| First | How can the United States secure its position as the unrivaled world leader in critical and emerging technologies — such as artificial intelligence, quantum information science, and nuclear technology — maintaining advantage over potential adversaries? |
| Second | How can we revitalize America’s science and technology enterprise — pursuing truth, reducing administrative burdens, and empowering researchers to achieve groundbreaking discoveries? |
| Third | How can we ensure that scientific progress and technological innovation fuel economic growth and better the lives of all Americans? |
The report was delivered on July 21, 2026, coinciding with the United States’ 250th anniversary — a deliberate symbolic alignment between national renewal and scientific renewal.
The Four Overarching Goals
Kratsios’s transmittal letter crystallizes the report’s ambition into four goals, with a unifying success metric:
“A decade from now, American researchers should look back at our work and say: ‘The vital questions I could not pursue then, I am free to pursue now.’”
| Goal | Core Principle |
|---|---|
| 1 | Prioritize the individual scientist over legacy institutions |
| 2 | Fundamentally change how research dollars are allocated, distributed, and assessed |
| 3 | Set clear scientific goals and build the industrial muscle to translate discovery into technological strength |
| 4 | Prepare the research enterprise for the AI revolution |
2. The Diagnosis: What’s Wrong with American Science
The report presents a multi-dimensional critique of the current U.S. scientific enterprise, organized around several interconnected failure modes.
2.1 The Productivity Paradox
Despite massive increases in research investment, scientific productivity — measured by breakthrough rate per dollar or per researcher — has declined:
| Indicator | Status |
|---|---|
| Breakthrough rate | Slowed despite massive biomedical funding increases since the 1990s |
| Drug approvals | Flatlined despite rising R&D expenditure |
| Research disruptiveness | Papers and patents are becoming less disruptive over time (Park et al., Nature, 2023) |
| Eroom’s Law in pharma | The number of new drugs approved per billion dollars of R&D spending has halved roughly every 9 years since 1950 |
Source cited: Bloom et al., “Are Ideas Getting Harder to Find?” American Economic Review (2020); Scannell et al., “Diagnosing the Decline in Pharmaceutical R&D Efficiency,” Nature Reviews Drug Discovery (2012)
2.2 Administrative Bloat
The report identifies bureaucratic overhead as a primary drag on scientific productivity:
- ~50% of researchers’ time is consumed by paperwork and administrative tasks
- Grant timelines have ballooned: Some grants now take nearly two years from submission to award — “almost as long as it took to design and produce the first Boeing 747”
- Federal and university bureaucracies have expanded, further reducing the share of funding that reaches actual science
- Indirect cost recovery (overhead charged by universities on federal grants) has been identified as a significant drain, supporting “administrative bloat” rather than scientific infrastructure
2.3 The “Incumbency Tax” and Weakened Meritocracy
The report argues that the current system systematically favors established researchers and incremental work:
| Problem | Description |
|---|---|
| Incumbency tax | Established researchers and institutions capture disproportionate funding, crowding out newcomers |
| Consensus-driven peer review | Review panels “gatekeep proposals by consensus, disincentivizing transformative ideas” |
| Weakened meritocracy | Selection does not rest purely on merit, with political considerations infiltrating funding decisions |
| Aging investigator pool | The average age of first-time NIH R01-equivalent awardees has risen from ~36 in 1980 to ~44 in recent years |
| “Science advances one funeral at a time” | Cited research (Azoulay et al., AER, 2019) showing that the death of a star scientist opens space for new ideas from outsiders |
2.4 The Reproducibility Crisis
The report directly addresses the replication crisis in science:
| Field | Replication Rate |
|---|---|
| Psychology | ~39% of studies replicated (Open Science Collaboration, Science, 2015) |
| Economics | ~61% of laboratory experiments replicated (Camerer et al., Science, 2016) |
| Preclinical cancer research | ~11% of landmark studies replicated (Begley & Ellis, Nature, 2012) |
| Estimated annual cost of irreproducible preclinical research | ~$28 billion (Freedman et al., PLoS Biology, 2015) |
2.5 The Shifting R&D Funding Landscape
The report highlights a fundamental structural shift in who funds American R&D:
| Metric | Data Point |
|---|---|
| Private sector R&D spending | ~$700 billion annually |
| Private vs. government + academia | Private sector spending is more than triple the combined spending of government and higher education |
| Industry share of national R&D | Nearly doubled from the 1950s to today |
| Federal R&D portfolio | ~$200 billion annually |
| Non-defense R&D concentration | Much of the increase is concentrated in the life sciences, while physical sciences and engineering have declined as a share of the federal research portfolio |
This evolution “made the pie bigger for everybody,” the report notes, “but it demands a corresponding adjustment to the nature of the Federal Government’s contributions.”
2.6 STEM Workforce Challenges
| Indicator | Trend |
|---|---|
| American citizens in post-graduate STEM | Declining proportion of U.S. citizens filling graduate STEM spots |
| Temporary visa holders with post-graduation commitments | A significant share of STEM doctorates are earned by international students, many of whom stay in the U.S. |
| International competition | Competitors are “channeling unprecedented resources into science and engineering, taking a whole-of-society approach to seize the high ground in strategic technologies” |
2.7 Loss of Industrial Capacity
Decades of offshoring have severed the link between American scientific discovery and domestic manufacturing:
- Manufacturing employment: Declined sharply over the past 40 years
- “Valleys of Death”: The gap between laboratory discovery and commercial-scale production has widened
- Tacit knowledge erosion: The hands-on craft and process knowledge that once translated discoveries into products has been dispersed globally
3. Chapter II: Revitalizing America’s Science & Technology Enterprise
Chapter II presents the report’s most granular reform proposals for how the federal government funds and organizes scientific research.
3.1 Five Strategic Reform Pillars
| Pillar | Key Proposals |
|---|---|
| Refocus on the Individual Scientist | — Expand portable graduate fellowships (modeled on NSF GRFP) — Back early-career independence — Scale long-horizon grants for top talent (modeled on NIH Director’s Pioneer Award) — Open alternative pathways beyond standard academia — Selection rests purely on merit |
| Diversify Funding Mechanisms | — “Golden tickets” empowering individual reviewers to champion ambitious proposals — Fast grants delivering decisions in days/weeks, not months/years — Prize challenges and advanced market commitments that pay for results — Regranting models delegating funding authority to scientists |
| Create New Institutional Models | — X-Labs: Agile, time-bound teams of professional scientists and engineers targeting specific bottlenecks — Advanced Research Projects Agencies (ARPAs): Individual program managers making bold bets — Curiosity-driven institutes: Long-horizon fundamental research |
| Reduce Bureaucratic Burdens | — Compress review cycles (currently up to 2 years) — Eliminate duplicative reporting — Rein in indirect cost recovery — Direct relief to early-career researchers |
| Institutionalize Continuous Improvement | — Empowered meta-science units in each federal science agency — Authority to run controlled experiments on review and funding mechanisms — Elevate program officers as “architects of the fields they help shape” |
3.2 The “Portfolio-Based Approach”
A recurring theme throughout Chapter II is that the federal government should treat its ~$200 billion annual R&D portfolio with the discipline of a sophisticated capital allocator:
“Federal research agencies should evaluate their own performance as capital allocators, test new ways of making grants, and construct portfolios with the intentionality of a serious investor.”
This represents a fundamental shift from the passive, proposal-driven model to an active, strategy-driven approach — where program officers exercise genuine discretion over scientific direction rather than simply administering peer review.
3.3 Comparison: Old Model vs. Proposed Model
| Dimension | Current Model (1945 Paradigm) | Proposed Model |
|---|---|---|
| Funding philosophy | Project-based, short-cycle grants | Multi-mechanism: fast grants + long-horizon grants + prize challenges |
| Review process | Consensus-driven peer review panels | Golden tickets + regranting + meta-science experimentation |
| Institutional focus | University departments, academic silos | X-Labs, ARPAs, curiosity-driven institutes, FROs |
| Grant duration | 3–5 year typical cycles | Fast (weeks) to long-horizon (10+ years) |
| Overhead/indirect costs | Escalating, poorly controlled | Capped, redirected to scientific infrastructure |
| Performance evaluation | Minimal self-assessment | Meta-science units, controlled experiments, continuous improvement |
4. Chapter III: Securing U.S. Dominance in Critical & Emerging Technologies
Chapter III shifts from internal reform to external competition, arguing that scientific leadership alone does not guarantee national strength — the U.S. must “tightly couple” discovery with domestic prototyping, manufacturing, and scaling.
4.1 Five Strategic Pillars for Technological Dominance
| Pillar | Key Proposals |
|---|---|
| Restore Permissionless Innovation | — Weigh benefits alongside risks in regulatory decisions — Extend reforms from nuclear, pharmaceuticals, and drones to other sectors — Regulatory sandboxes for testing new technologies under controlled conditions — Streamline permitting, reducing the “costs of inaction” |
| Open Federal Infrastructure | — Broaden industry access to DOE national labs, NASA centers, and DOW facilities — Streamline CRADAs (Cooperative Research and Development Agreements) — Leverage Other Transaction Authority (OTA) for private-sector engagement — Joint public-private investments in cutting-edge equipment |
| Strengthen Public-Private Partnerships | — Expand agency-adjacent foundations — Focus SBIR/STTR programs on strategic capabilities — Joint centers among industry, academia, and federal facilities — Scale industry Ph.D. and postdoc fellowships for cross-sector talent flow |
| Organize Pre-Competitive Consortia & Grand Challenges | — Moonshot-scale missions for national priorities — Industry consortia to break shared engineering bottlenecks — Modeled on EUV LLC (semiconductor lithography consortium) and the Human Genome Project |
| Use States as Laboratories | — State-led regulatory and economic experimentation — Partner with fastest-moving jurisdictions — Spread successful models across all counties and states — Examples cited: Arizona’s autonomous vehicle framework, Utah’s regulatory sandbox, Wyoming’s crypto-friendly laws |
4.2 Key Case Studies Referenced
| Case Study | Significance |
|---|---|
| EUV LLC (Extreme Ultraviolet Lithography) | A pre-competitive consortium of semiconductor companies and national labs that solved the lithography bottleneck, enabling advanced chip manufacturing. The report presents this as the model for future industry consortia. |
| Human Genome Project | Demonstrated that the federal government could marshal scientific effort at a scale no single institution could match; delivered a $141 return for every $1 invested (Battelle study, 2011). |
| Apollo Program | The archetype of a grand challenge: government-articulated goal, mobilization of national resources, technological spillovers across the entire economy. |
4.3 The Strategic Competition Dimension
The report frames technological competition in explicitly geopolitical terms:
“As our competitors race us to capture the value chain of strategic technologies, including with tactics we would never countenance, we can no longer assume that the fruits of American science will accrue to our own people.”
- Competitor tactics identified: Whole-of-society approaches to R&D, state-directed industrial policy, and aggressive talent acquisition
- Domestic response: Combine scientific excellence with industrial capacity — not one without the other
- “Fighting in our own arena”: An explicit argument that the U.S. should leverage its unique strengths (federalism, private sector dynamism, regulatory flexibility) rather than imitate competitors’ models
5. Chapter IV: Ensuring Science & Technology Benefit All Americans
Chapter IV addresses the distributional dimension of scientific progress — arguing that the economic returns of discovery must accrue to American workers and communities, not just to consumers and shareholders.
5.1 Four Strategic Pillars for Broad-Based Prosperity
| Pillar | Key Proposals |
|---|---|
| Integrate Hands-On Training | — Embed practical technical training and externships into all STEM curricula — Let hands-on experience and industry credentials count toward degrees — Reform accreditation, admissions, and tenure to reward real-world technical work — “Technology is encoded not just in papers and patents, but in the tacit knowledge passed from mentor to mentee” |
| Open Scientific Careers Beyond Academia | — National fellowships for skilled craftspeople — Practitioner-in-residence programs embedding machinists and technicians alongside Ph.D. researchers — Portable industry-recognized credentials in advanced manufacturing and lab techniques — Connect hobbyists and tinkerers in rural communities to formal research |
| Modernize Apprenticeships | — Extend registered apprenticeships into science and technology fields — Pay-for-performance funding models — Scale Workforce Pell Grants — Community colleges as regional hubs of scientific and technical talent |
| Build Dense Local Innovation Clusters | — Expand regional innovation hubs, manufacturing institutes, and defense industrial base centers — Co-locate research and production — Reshore advanced manufacturing — Restore feedback loops between researchers, engineers, and skilled technicians |
5.2 The Marriage of Science and Craft
A distinctive thread in Chapter IV is the elevation of tacit knowledge and skilled trades to equal standing with academic credentials:
“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.”
This philosophy draws on Michael Polanyi’s concept of “tacit knowledge” — understanding that cannot be fully codified in writing but must be learned through practice and apprenticeship. The report argues that the U.S. has systematically undervalued this dimension of technological capability.
6. Chapter V: A New Golden Age — The AI Transformation
Chapter V is the most forward-looking section of the report, describing a vision for how artificial intelligence will transform the practice of science itself — and what institutional changes are needed to harness this transformation.
6.1 The Core Argument
“America stands at the cusp of a revolution in science, in which AI will accelerate discovery, multiply human cognitive capabilities, and unlock solutions to some of our greatest challenges. But ‘AI for science’ will still find itself subject to the frictions and inefficiencies of human institutions.”
The report argues that AI’s transformative potential for science is comparable in scale to the printing press or the research university — but that legacy institutions, designed for human-paced discovery, will become the binding constraint unless proactively restructured.
6.2 Five Strategic Pillars for AI-Enabled Science
| Pillar | Key Proposals |
|---|---|
| Launch and Scale the Genesis Mission | — America’s flagship AI-for-science initiative — Integrate supercomputers, AI models, scientific instruments, and datasets across national laboratories — Goal: Double the productivity and impact of U.S. science within a decade — Target cross-cutting problems where breakthroughs unlock entire branches of downstream discovery |
| Institutionalize Gold Standard Science | — Enforce reproducibility, transparency, data sharing, and falsifiability across all federally funded research — Execute through the Restoring Gold Standard Science Executive Order (EO 14303) — Create a trusted foundation for AI-powered discovery — “AI operating on a flawed knowledge base will only entrench bad science” |
| Build Verification Infrastructure at Scale | — “While the cost of generation has decreased exponentially, the cost of verification has not” — AI-enabled verification systems, open standards, continuous replication mechanisms — Machine-auditable replication packages — Reward those who replicate or disprove influential results |
| Accelerate Autonomous Experimentation | — Closed-loop autonomous laboratories (robotics + AI) — “Collapse discovery timelines by orders of magnitude” — Enable science at “truly industrial scale” — Federal R&D to build the scientific equipment industrial base |
| Experiment with AI-Native Scientific Institutions | — Faster, more open forms of scientific publication — More granular credit attribution (blockchain-based systems) — New market mechanisms directing resources to high-impact problems — Transition from human-paced to AI-paced discovery infrastructure |
6.3 The Liquid Tensor Experiment & Prime Number Theorem: AI-Verified Mathematics
The report provides a compelling case study in AI-enabled verification:
| Milestone | Details |
|---|---|
| Liquid Tensor Experiment (2020–2022) | Formalizing a single theorem in condensed mathematics into machine-checkable code consumed nearly two years of effort from expert practitioners |
| Prime Number Theorem — Traditional Approach (2024–2025) | An ambitious project with 20+ collaborators worldwide made intermediate progress over 18 months but remained stuck on core difficulties in complex analysis |
| Prime Number Theorem — AI Approach (September 2025) | A startup using AI completed the project in three weeks, spanning 1,100 formally verified theorems and definitions |
Implication: When checking an answer is easier than finding one — the classic asymmetry of mathematics — AI transforms not just how fast we discover, but who can participate and what problems become tractable.
6.4 Rethinking Scientific Publication
The report launches a sustained critique of the journal system:
| Problem | Description |
|---|---|
| Scale mismatch | Journals designed for hundreds of researchers now serving 9 million full-time researchers worldwide, publishing millions of articles across tens of thousands of journals |
| Artificial scarcity | Journals reward secrecy over open collaboration |
| Anonymous gatekeeping | A small number of unpaid reviewers with vested interests determine what counts as legitimate science |
| Narrative bias | The format rewards polished narratives over honest accounts of the research process |
| Suppressed findings | Null findings, failed experiments, and methodological details rarely reach publication |
| Goodhart’s Law risk | When AI can generate plausible research at industrial scale, metrics like papers published, citations accumulated, and impact factors “will all fall to Goodhart’s Law as gameable targets” |
Proposed future: Shorter, more frequent outputs (datasets, code, preliminary findings, methodological notes); dynamic papers that update as data changes; public peer review conducted in the open; machine learning community norms (rapid replication, social media debate) extended across disciplines.
6.5 A Vision: The Agent-Based Scientific Economy
In its most speculative section, the report sketches a future where AI agents participate as autonomous actors in a decentralized scientific economy:
| Component | Description |
|---|---|
| Granular credit attribution | Blockchain-based immutable records of every contribution — datasets uploaded, analyses run, hypotheses proposed — timestamped and linked to creators |
| Decentralized Autonomous Organizations (DAOs) | Communities that pool resources and fund research directly, bypassing traditional institutional gatekeepers |
| Prediction polls | Augmented with proper scoring feedback and statistical aggregation, forecasting scientific developments |
| Continuous market-mediated discovery | A funder posts a million-dollar bounty for a validated therapeutic target → an agent notices a promising lead → posts a smaller bounty for replication → other agents contract autonomous laboratories → cryptographically signed results → smart contracts release funds as milestones are verified |
| Cloud laboratories | Robotic facilities synthesizing molecules, running assays, returning results without human intervention — becoming “factories of this economy” |
“In such a world, experimental information becomes a tradeable commodity, and price mechanisms replace slow institutional coordination.”
The report is careful to note: “None of this exists today in a mature form, but the pieces are emerging separately.”
6.6 “As We May Build” — The Closing Vision
The report’s final section, titled after Vannevar Bush’s 1945 essay “As We May Think,” lays out a long-term technological horizon:
| Domain | Vision |
|---|---|
| Materials science | Metamaterials that bend light in unprecedented ways, programmable matter, self-assembling structures — “the progression from semiconductor fab to molecular assembler may take only decades” |
| Biomedicine | Engineering cells as precisely as circuits, programming immune systems to hunt malignancies, designing therapeutics atom by atom |
| Manufacturing revival | More than $1 trillion in investment commitments secured in the first year of the Trump Administration for advanced manufacturing and physical-world technology companies |
| Craft revival | Precision bearings toleranced to millionths of an inch, carbon nanofibers for spacecraft, semiconductor crystals of “inhuman purity” |
| Infrastructure for discovery | “Trillions of AI agents running experiments, testing conjectures, and surfacing insights across every scientific domain” |
7. The Genesis Mission & Gold Standard Science
7.1 The Genesis Mission
The Genesis Mission, launched via Executive Order 14363 (November 2025), is the Trump Administration’s flagship AI-for-science initiative:
| Element | Details |
|---|---|
| Objective | Double the productivity and impact of U.S. science within a decade |
| Mechanism | Integrate supercomputers, AI models, scientific instruments, and datasets across national laboratories |
| Lead agency | Department of Energy (DOE) |
| Initial collaborators | 24 organizations announced collaborative agreements (December 2025) |
| Transformational AI Models Consortium | DOE National Laboratory Program, announced August 2025 (LAB 25-3560) |
| Strategic focus | Cross-cutting problems where breakthroughs unlock entire branches of downstream discovery and where AI can transform the practice of science itself |
7.2 Gold Standard Science (Executive Order 14303)
The “Restoring Gold Standard Science” Executive Order (May 2025) establishes:
| Requirement | Purpose |
|---|---|
| Reproducibility | All federally funded research must be reproducible |
| Transparency | Data, methods, and analysis must be openly shared |
| Data sharing | Mandated data deposit and access protocols |
| Falsifiability | Research must generate testable, falsifiable claims |
| NIH Replication Initiative | Agency-wide initiative to elevate replication and reproducibility studies, identifying critical research and infrastructure needs |
The report frames Gold Standard Science as not merely a quality-control measure but as a prerequisite for AI-enabled discovery: AI models trained on flawed, irreproducible research will “only entrench bad science.”
8. FY 2028 R&D Budget Priorities Memorandum (NSTM-5 / M-26-16)
The Annex to the report is a formal memorandum co-signed by OSTP Director Kratsios and OMB Director Russell Vought, issued July 21, 2026, providing binding budget guidance to all federal departments and agencies for FY 2028 formulation.
8.1 Nine Priority Areas
| # | Priority Area | Key Directives |
|---|---|---|
| (i) | Rebalance R&D portfolios toward foundational research | — Increase the share of foundational research relative to later-stage development — Prioritize physical sciences (physics, chemistry, materials science, space science), computer science, and supporting engineering and mathematical disciplines — Prioritize foundational biological sciences over applied life sciences |
| (ii) | Advance national S&T missions | — AI, quantum information science, semiconductors, advanced communications, robotics, advanced manufacturing, nuclear fission and fusion, space systems |
| (iii) | Apply AI and emerging technologies to accelerate research | — Integrate AI tools across the scientific workflow — Build the infrastructure for AI-enabled science |
| (iv) | Expand R&D infrastructure for broader ecosystem use | — Open federal laboratory infrastructure to industry and academic users — Joint investments in cutting-edge equipment |
| (v) | Translate scientific advances into stronger regional ecosystems | — Regional innovation hubs, manufacturing institutes — Reshoring advanced manufacturing |
| (vi) | New funding mechanisms and institutional models | — Golden tickets, fast grants, prize challenges, ARPAs, X-Labs, FROs — Diversify beyond consensus-driven peer review |
| (vii) | Better ways to identify and develop scientific talent | — Portable fellowships, early-career independence, alternative career pathways — Merit-based selection; expand beyond traditional academic pipeline |
| (viii) | Rigorously study, evaluate, and improve how federal science is funded | — Stand up meta-science units in each agency — Run controlled experiments on funding mechanisms — Evaluate agency performance as capital allocators |
| (ix) | Integrate federal R&D into the broader S&T enterprise | — Closer partnerships with private sector, philanthropy, and state/local governments — Rebalance federal contributions given industry’s dominant funding role |
8.2 Specific Physical Sciences Priority
The memorandum emphasizes a structural rebalancing toward fields that have been underweighted:
“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 directed to note the R&D character classification of proposed activities as a percentage of their portfolios and identify specific programs shifting toward earlier-stage work. Where agencies propose significant expansion of later-stage development, they must justify why such activities would not occur absent federal support.
8.3 Specific Physical Sciences Sub-Fields Prioritized
- Condensed matter and quantum materials physics (correlated, magnetic, and topological states)
- Photonics
- Advanced materials science
- Space science
- Chemistry
- Computer science fundamentals
- Mathematical disciplines supporting engineering
9. Key Data Points at a Glance
| Data Point | Value | Source Context |
|---|---|---|
| U.S. private sector annual R&D spending | ~$700 billion | >3x government + higher education combined |
| Federal annual R&D portfolio | ~$200 billion | Across all agencies |
| Industry share of national R&D funding | Nearly doubled since 1950s | Structural shift in funding landscape |
| Researcher time lost to administration | ~50% | Paperwork, compliance, reporting |
| Maximum grant timeline | ~2 years from submission to award | “Almost as long as the first Boeing 747” |
| Average age of first NIH R01 awardee | ~36 (1980) → ~44 (recent) | Aging investigator pool |
| Psychology replication rate | ~39% | Open Science Collaboration, 2015 |
| Economics replication rate | ~61% | Camerer et al., 2016 |
| Preclinical research irreproducibility cost | ~$28 billion/year | Freedman et al., 2015 |
| Human Genome Project ROI | $141 per $1 invested | Battelle, 2011 |
| Global full-time researchers | ~9 million | Publishing millions of articles across tens of thousands of journals |
| AI-completed Prime Number Theorem formalization | 1,100 verified theorems in 3 weeks | vs. 18+ months for 20-person traditional effort |
| Genesis Mission collaborators | 24 organizations (initial) | Announced December 2025 |
| Advanced manufacturing investment commitments | >$1 trillion | First year of Trump Administration |
| American agricultural productivity | Output tripled with ~1/4 the labor | Since 1940s |
| U.S. life expectancy gain | >10 years longer | Since Bush’s 1945 report |
| Manufacturing employment trajectory | Steep decline over 40 years | BLS data |
| Pharma R&D efficiency (Eroom’s Law) | New drugs per $B R&D halved ~every 9 years | Since 1950 |
10. Strategic Implications & Analysis
10.1 A Paradigm Shift in Federal Science Policy
Science: A New Golden Age is not a marginal adjustment to existing policy. It represents the most comprehensive proposal to restructure the American scientific enterprise since 1945. The report’s core thesis — that the institutions built on Bush’s Endless Frontier are now the binding constraint, not the enabling foundation — carries implications across every dimension of federal R&D.
10.2 The “Capital Allocator” Model
Perhaps the most consequential organizational proposal is the redefinition of federal science agencies as active capital allocators rather than passive grant administrators. If implemented, this would:
- Give program officers significantly expanded discretion over scientific direction
- Introduce portfolio-level performance metrics
- Shift funding from consensus-driven incremental work toward high-risk, high-reward proposals
- Create internal competitive pressure for funding mechanism innovation
10.3 The Physical Sciences Rebalancing
The explicit directive to rebalance toward physical sciences and engineering — reversing a decades-long shift toward the life sciences — signals a strategic reorientation. The report frames this in national security terms: the technologies that will define geopolitical competition (AI, quantum, advanced manufacturing, nuclear, space) all depend on foundational physical sciences research.
10.4 The AI Transformation Thesis
The report’s vision of an “agent-based scientific economy” — while speculative — represents a genuine attempt to think through the institutional implications of AI, rather than merely funding more AI research within existing structures. The emphasis on verification infrastructure, Gold Standard Science, and new publication models all flow from the premise that AI will generate knowledge at a scale and speed that legacy institutions cannot handle.
10.5 International Competitive Dynamics
The report is notable for its explicit framing of science policy as a domain of geopolitical competition. The diagnosis that competitors are using “tactics we would never countenance” — state-directed industrial policy, aggressive intellectual property acquisition, whole-of-society technology strategies — reflects a significant hardening of the U.S. posture. The proposed response, however, leans into distinctively American strengths (federalism, private sector dynamism, immigration of talent) rather than imitation.
10.6 Implementation Challenges
Several elements of the report face significant implementation hurdles:
| Challenge | Description |
|---|---|
| Congressional appropriations | Many proposals require legislative action or reallocation of appropriated funds across entrenched interests |
| University resistance | Indirect cost reform, tenure reform, and the shift toward non-academic performers threaten the financial model of research universities |
| Cultural inertia | The peer review consensus model is deeply embedded in scientific culture; “golden tickets” and prize challenges face skepticism |
| Coordination complexity | Integrating DOE, NIH, NSF, NASA, DOD, and other agencies around shared missions requires unprecedented interagency coordination |
| Verification economics | Building continuous, low-cost verification infrastructure at the scale of global scientific output is an unsolved economic and technical challenge |
10.7 Historical Significance
The report self-consciously positions itself as the successor to Science: The Endless Frontier — released 81 years after the original, on the occasion of the nation’s 250th anniversary, with an explicit parallel between FDR’s 1944 letter to Bush and Trump’s 2025 letter to Kratsios. Whether it achieves comparable historical influence will depend on implementation — but as a diagnostic document and strategic blueprint, it represents the most ambitious rethinking of American science policy in the post-war era.
Methodology & Source Notes
- Primary source: Science: A New Golden Age — A Report to the President, by Michael Kratsios, Director of the White House Office of Science and Technology Policy, July 2026 (125 pages)
- Annex source: Fiscal Year 2028 Administration Research and Development Budget Priorities Memorandum (NSTM-5 / M-26-16), co-signed by OSTP Director Kratsios and OMB Director Vought, July 21, 2026
- Executive Orders referenced: EO 14363 (Genesis Mission, November 2025), EO 14303 (Restoring Gold Standard Science, May 2025)
- Key precursor document: Vannevar Bush, Science: The Endless Frontier (1945)
- All data points, statistics, and quotations are sourced directly from the report unless otherwise noted
- This article is designed to be a complete, standalone reference that does not require access to the original document
