Ecosystem Forces: The External Conditions Shaping What Research Is Possible

European defence priorities have surged; dual-use research has gained legitimacy and funding; sanctions have fragmented international collaboration networks built over decades. For Nordic countries specifically, the war has accelerated a reassessment of strategic science dependencies — in energy, materials, and security-adjacent technology. The long-term consequence for Denmark is a likely reorientation of priority research areas toward resilience, defence, and energy security, with foundation capital increasingly drawn into that frame.

Sources: Widely reported. The Nordic and Danish-specific implications are analytical inference from the general pattern rather than directly sourced; no systematic study of the consequences for Danish research priorities had been published at time of writing.

Restrictions on researchers, equipment, and data flows are tightening across a growing range of fields. Semiconductor and AI research are increasingly treated as national security assets. The decoupling is now operational rather than merely rhetorical: US legislation restricts certain AI chip exports; Chinese postdoctoral researchers in US institutions have faced prosecution and deportation under China Initiative successor programmes; there is documented researcher flight from US institutions among Chinese-born scientists who no longer feel secure. International scientific collaboration is bifurcating into blocs — not uniformly, but specifically in quantum computing, advanced materials, and AI hardware where political stakes are highest. European institutions are caught between maintaining both relationships while being pressured to choose sides.

The 2025 trade data makes the decoupling empirically concrete in ways that earlier analyses could not. McKinsey Global Institute's tracking of actual trade flows across economies covering 90% of global goods trade shows US–China bilateral trade falling by around 30%, with the United States replacing roughly two-thirds of the gap through alternative suppliers — including India for smartphones (reaching roughly 40% of what China had supplied) and ASEAN economies for laptops and electronics. The average effective US tariff rate peaked at approximately 22% in April 2025 — the highest in roughly a century — before partial negotiations brought it to about 15% by year-end. Simultaneously, AI-related hardware trade (semiconductors, data centre equipment, networking hardware) grew by roughly 40% globally and accounted for about a third of all global trade growth — making the most politically contested goods category simultaneously the most economically dynamic. China, increasingly cut off from the most advanced US chips, expanded AI-related trade by roughly 16% through domestic sourcing while the US–Taiwan–South Korea–ASEAN supply chain expanded sharply in the other direction. European ultraviolet lithography machines for advanced chipmaking remain critical to both sides, making ASML — based in the Netherlands — a geopolitically exposed pivot point that has imposed its own export controls under US pressure.

Sources: Widely reported across policy and science press. The field-specific dynamics (quantum, semiconductors, AI hardware) are well-documented in policy literature; the European positioning is analytical inference. For the most comprehensive recent analysis of China's techno-industrial strategy — five channels, 18 instruments, 15th FYP — see Gerard DiPippo, Jonathon Sine & Benjamin Lenain, China's Techno-Industrial Strategy in the Xi Era: Producing Under Pressure (RAND, 2026). 2025 trade data from McKinsey Global Institute, Geopolitics and the Geometry of Global Trade: 2026 Update (March 2026).

The post-Cold War assumption of borderless knowledge exchange is eroding. Which problems get worked on jointly, and by whom, is increasingly political. The international collaboration infrastructure — joint institutes, shared databases, cross-border research programmes — was built under conditions that no longer fully hold. This fragmentation is not uniform: the UK's association with Horizon Europe (January 2024) is a partial reversal of the Brexit-era rupture, and Ukrainian scientists are being integrated into EU frameworks on expedited timelines. But the overall direction of travel remains toward more politically mediated access to the global science system. The implications for small open research nations like Denmark, whose scientific output is heavily dependent on international collaboration, are significant and largely unaddressed at the policy level.

Survey evidence now quantifies where global leaders expect this trajectory to lead. The WEF's 2026 Global Risks Perception Survey asked 1,300+ respondents to characterise the global political environment for cooperation over the next 10 years: 68% selected a "multipolar or fragmented order in which middle and great powers contest, set, and enforce regional rules and norms"; only 6% expected reinvigoration of the US-led, rules-based international order. McKinsey Global Institute's trade analysis provides a concrete economic analogue: the geopolitical distance of goods trade — a measure of how politically distant trading partners are — declined at a −1.2% annualised rate in 2024–25, accelerating a trend underway since 2017, as goods increasingly flow between politically aligned partners. If knowledge flows tend to follow goods flows, and goods flows are visibly re-aligning along bloc lines, then the structure of international scientific collaboration is under the same pressure — not uniformly or immediately, but systematically.

Sources: The general fragmentation trend is widely reported. UK-Horizon association is confirmed from official EU and UKRI sources. The specific exposure of small open research nations is analytical inference; the Nielsen & Qiu framing of science as a set of social processes that can be disrupted as well as designed is relevant background.¹ Survey data from World Economic Forum, The Global Risks Report 2026 (January 2026); trade data from McKinsey Global Institute, Geopolitics and the Geometry of Global Trade: 2026 Update (March 2026).

Mobility of researchers across borders — for training, collaboration, and career development — has been a structural feature of the global science system for decades. That mobility is now being actively contested. Restrictions on Chinese and Iranian-born researchers working in US institutions have created legal risk for international scientific collaboration in certain fields. The secondary talent displacement — researchers self-censoring international collaboration to avoid scrutiny — is harder to measure but likely larger than direct prosecutions suggest. In the other direction, France, Germany, the Netherlands, and others have launched active recruitment campaigns specifically targeting researchers displaced from US institutions. The aggregate effect is that access to global talent is increasingly a policy instrument, not a background condition — which reshapes the competitive position of institutions differently depending on their national regulatory environment.

Sources: The US restrictions on Chinese postdocs and researchers are widely reported and documented in academic union and legal reporting. European recruitment campaigns (France's Atrae programme) are widely reported. The self-censorship dynamic is analytical inference from the documented pattern.

The World Economic Forum's 2026 Global Risks Report — drawing on surveys of 1,300+ experts across academia, business, government, and international organisations — ranks geoeconomic confrontation as the single most severe short-term global risk, selected by 18% of respondents as most likely to present a material crisis in 2026. The core concept: trade, finance, technology access, and supply chains are no longer primarily economic relationships governed by market logic — they are instruments of statecraft, deployed for strategic advantage. This distinguishes the current period from earlier episodes of trade friction; the goal is not merely favourable terms but control over the levers of interdependence itself.

The concrete manifestations are substantial. US tariff rates reached their highest level since World War II in 2025, peaking at approximately 22% average effective rate in April before partial negotiations brought it to about 15% by year-end. China responded with rare earth export restrictions and retaliatory tariffs on US agricultural products. Multiple economies implemented investment screening in strategic sectors. Export controls on advanced semiconductors, AI chips, and chipmaking equipment have created a de facto technology frontier running through Taiwan, South Korea, and the Netherlands. McKinsey Global Institute's analysis of actual 2025 trade flows shows US–China bilateral trade falling by around 30%, with ripple effects across global supply chains as firms rerouted to third countries. UNCTAD documents that global geopolitical risk indicators reached their highest recorded level since the onset of the Ukraine war by early 2026, driven by Middle East escalation that sent oil surging over 60% in days and triggered the biggest combined sell-off in stocks and bonds since 2022.

The implications for science are structural. Research is increasingly treated as a strategic asset to be protected rather than a global common to be shared. Collaboration in advanced fields — quantum computing, semiconductors, AI hardware, synthetic biology — is subject to export control logic that treats knowledge transfer as equivalent to weapons transfer. The fields in which states are most actively seeking dominance are precisely those most central to the current research frontier. Institutions whose scientific strategies depend on open international collaboration in technically sensitive areas are exposed in ways they have not previously had to plan for.

Sources: World Economic Forum, The Global Risks Report 2026 (21st edition, January 2026); McKinsey Global Institute, Geopolitics and the Geometry of Global Trade: 2026 Update (March 2026); UNCTAD, Trade and Development Foresights 2026: Global Economy Faces a Geopolitical Challenge (April 2026); EU Council General Secretariat ART, Prospects for 2026: What Others Are Saying (January 2026). The science-specific implications are analytical inference from the documented pattern.

AI is changing what is computable, automatable, and institutionally tractable in research. In certain fields — structural biology, materials science, drug discovery — it compresses discovery timelines and changes the economics of research organisations: a small team with the right tools can now do what required a large lab a decade ago. AlphaFold was solved by 19 people.² A large empirical study (Evans et al., Nature 2026, 41.3 million papers) gives this a precise shape: scientists using AI publish 3× more and receive nearly 5× more citations. The countervailing finding is that AI adoption simultaneously shrinks the diversity of topics studied by 4.6% and reduces cross-scientist engagement by 22%, as tools pull researchers toward data-rich, benchmark-legible areas. The productivity gain is real; so is the homogenisation.

The scale of AI-driven investment has become visible in goods trade in ways that underscore the industrial reality of the shift. McKinsey Global Institute's analysis of 2025 global trade flows shows AI-related hardware — semiconductors, servers, graphics cards, networking equipment, and data centre construction materials — growing by roughly 40% and accounting for approximately a third of all global trade growth. The United States added roughly half of the world's new data centre capacity in 2025, making it the single largest driver of global import growth as a consequence. The WEF's 2026 Global Risks Report registers the downstream concern: "Adverse outcomes of AI technologies" made the sharpest increase of any risk in the 10-year outlook, rising from #30 to #5 — ahead of natural resource shortages and state-based armed conflict. This places AI in an unusual position: simultaneously the fastest-growing trade category, the most geopolitically contested technological domain, and the risk most sharply upgraded by expert assessment. Institutions making long-run bets on AI-enabled science are making those bets in that compound environment.

Sources: The AlphaFold team-size point is documented.² The broader claims about AI's effects on research economics are analytical, widely discussed in the metascience and science policy press. The productivity/homogenisation finding is from Evans et al., "Artificial intelligence tools expand scientists' impact but contract science's focus," Nature (January 2026). Trade data from McKinsey Global Institute, Geopolitics and the Geometry of Global Trade: 2026 Update (March 2026); risk-ranking data from World Economic Forum, The Global Risks Report 2026 (January 2026).

CRISPR, cell therapy, and gene drives are maturing from research tools into deployment-scale technologies. This creates new urgency around biosecurity as a research governance question and new dual-use risks that existing institutional structures are poorly designed to manage. Foundation capital — particularly NNF — is deeply invested in this space, which gives the Nordic ecosystem unusual leverage but also unusual exposure to the governance risks that come with it.

Sources: The maturation of CRISPR and cell therapy is widely reported. The NNF investment focus on life sciences is documented in its annual reports.³

Long timelines, but with potentially transformative consequences for specific problem classes — cryptography, materials simulation, optimisation. Currently at the stage where national and bloc-level investment is high and geopolitically motivated. The research infrastructure being built now will shape what is tractable in 10–15 years. Denmark has made unusually large bets relative to its size: the QuNorth initiative (€80M) is building a world-class quantum computer at DTU, and 55 North is a €300M quantum-focused investment fund anchored by EIFO and Novo Holdings — reported to be the world's largest dedicated quantum VC vehicle.

Sources: Widely reported. The geopolitical dimension is particularly visible in EU and US policy documents on strategic technology. QuNorth and 55 North figures are from Danish research and business press.

Anthropic, DeepMind, OpenAI, Meta AI, and their peers are not research institutions in any formal governance sense — but they are producing frontier science across multiple fields simultaneously, attracting researchers who would otherwise work in universities or national labs, and operating entirely outside the governance frameworks that apply to publicly funded research. The OpenAI Foundation holds a 26% stake in a company valued in the hundreds of billions — an endowment structure that, if realised, would rival the largest research foundations in the world. These organisations represent a new institutional form for doing science at scale: faster, less constrained by peer-review norms, more secretive in some respects, more open in others. Their growing role in structuring what research gets done and by whom is almost entirely unaddressed in metascience and science policy literature. That gap is itself a significant fact about where the conversation is.

Sources: The institutional analysis is synthesised from widely reported developments. OpenAI Foundation's stake and company valuation are from press reporting; actual figures fluctuate with market conditions. The governance gap — AI labs operating outside standard research governance — is analytical inference from publicly observable features of these organisations. For a first-person strategic account of how a lab founder sees the relationship between AI capability and science policy across biomedicine, energy, economic development, and governance, see Dario Amodei, "Policy on the AI Exponential" (June 2026).

The largest sustained gravitational pull on research funding priorities in the current period. Carbon capture, clean energy, adaptation, and climate modelling are drawing capital from governments, foundations, and industry at a scale that is reshaping which fields have momentum and which institutions are well-positioned. For Denmark, the connection is direct: in 2024, Danish industrial foundations distributed DKK 4.1 billion to nature and climate causes — the second-largest giving category after health⁴ — and national commitments to 70% emissions reduction create additional policy-driven research pressure.

A significant shift is underway in how expert opinion and policymakers are weighting environmental risks against geopolitical and economic ones in the near term. The WEF's 2026 Global Risks Perception Survey — 1,300+ experts — shows a notable short-term decline in environmental risk rankings: Extreme weather events fell from #2 to #4; Critical change to Earth systems fell from #7 to #24; Biodiversity loss and ecosystem collapse fell five positions. The EU Council's synthesis of 35 major global outlook reports describes the climate agenda as having "become a geopolitical casualty" — with both EU and US governments pulling back from commitments under pressure from military spending demands, economic austerity, and domestic politics. In the long run (10-year horizon), environmental risks still dominate — Extreme weather events retains the #1 position. But the near-term institutional environment for climate research funding is measurably shifting toward geopolitical and security priorities, and the gap between what the science says is urgent and what governments are currently prioritising is widening rather than closing.

Sources: The DKK 4.1bn figure is from Fondenes Videnscenter "Fondsbevillinger 2024."⁴ The general direction of foundation and government funding is widely documented. The short-term deprioritization finding is from World Economic Forum, The Global Risks Report 2026 (January 2026); the "geopolitical casualty" framing from EU Council General Secretariat ART, Prospects for 2026: What Others Are Saying (January 2026).

Slower-moving than the climate conversation but structurally similar: creates demand for long-horizon, field-based, integrative science that current institutions handle poorly. Funding is growing but remains fragmented across government, NGO, and foundation sources. Increasingly linked to climate in combined "nature and climate" funding categories — both in Denmark and internationally.

Sources: The structural similarity to climate is analytical. The Danish combined category is documented in Fondenes Videnscenter data.⁴

New materials, grid systems, storage, and nuclear (fusion and small modular reactor revival). Creating industrial-policy-funded research at a scale not seen since the space race. The transition is fast enough that the relevant research is as much about deployment and systems integration as fundamental discovery — a different institutional requirement from standard academic science, favouring engineering-oriented labs and mission-driven organisations over traditional university research groups.

Sources: Widely reported. The institutional implications — favouring mission-driven over university research — are analytical, consistent with the broader new institutions literature.

The Biden-era IRA and CHIPS Act were the most ambitious US research and industrial policy in decades — redirecting capital at scale toward clean energy, semiconductors, and advanced manufacturing, and establishing new institutional forms (Natcast, national technology centres) around those priorities. Since January 2025, both programmes have been substantially reversed: IRA disbursements were frozen on January 20, 2025; Natcast — the nonprofit research vehicle created under CHIPS to manage a committed $7.4 billion — was defunded in August 2025.

The reversal has itself become a force. What was presented as a generational industrial policy template proved highly vulnerable to electoral change. European and Asian governments are now accelerating their own responses partly to fill the gap — and institutions that built their strategies around policy-driven US capital now face structural uncertainty. The episode is a significant data point for any institution building on the durability of government-directed research funding.

A notable counterpoint within the same period: NSF's $1.5 billion X-Labs initiative (announced June 2026) explicitly funds independent research organisations on FRO-like terms — three to seven-year timeframes, defined problem scopes, grants scaling from $1.5M to $50M — representing deliberate institutional innovation within federal science policy even as the broader IRA/CHIPS reversal continues.

Sources: Widely reported. The IFP and FAS have produced detailed analyses of the broader US science contraction.⁵ The IRA freeze and Natcast defunding were covered extensively in the US science press in 2025. NSF X-Labs: Lucas Laursen, "NSF Experiments With New Kind of Science Funding," IEEE Spectrum (June 4, 2026).

European strategic autonomy as an explicit research governance principle. Horizon Europe and its successors are increasingly shaped by geopolitical and industrial-policy logic rather than purely scientific merit criteria. For small member states, this creates both opportunity (access to larger funding pools) and constraint (research priorities set in Brussels rather than Copenhagen).

Sources: Widely reported. The implications for small member states are analytical.

China has operated the most sustained and largest-scale state-directed science and technology programme in the world for over a decade. The 15th Five-Year Plan (2026–2030) deepens a strategy built across five channels and 18 distinct policy instruments — spanning domestic production mandates, state investment vehicles, talent acquisition, standards-setting, and international technology transfer. The goal is not merely to catch up with Western science but to establish technological leadership in specific fields: semiconductors, AI, quantum computing, advanced materials, biotechnology, and clean energy. China's STEM graduate output now substantially exceeds the United States and Europe combined. Its publication volume has overtaken the US in several fields by quantity, with rapid growth in citation impact.

These innovation-capacity indicators, however, likely overstate China's actual S&T power. Jeffrey Ding's systematic decomposition of the Global Innovation Index shows that China's average global ranking on innovation-capacity indicators (around 14th) diverges sharply from its ranking on diffusion-capacity indicators — how widely new technologies actually spread across firms, industries, and the broader economy (around 47th). That gap of over 30 places is the largest of any country studied, and has persisted consistently across 2014, 2017, and 2020 data. The parallel with the Soviet Union is direct: the USSR led on innovation indicators throughout the 1950s–70s and was widely assessed as a rising S&T superpower, but its failure to diffuse innovations economy-wide preceded its economic stagnation and eventual collapse. This does not mean China's investment is inconsequential — it creates genuine competitive pressure in specific frontier domains — but it does mean that conventional assessments of China's S&T rise, built from the same innovation-centric indicators, may significantly inflate its overall trajectory.

Sources: Gerard DiPippo, Jonathon Sine & Benjamin Lenain, China's Techno-Industrial Strategy in the Xi Era: Producing Under Pressure (RAND, 2026) for the five-channel, 18-instrument structure. The diffusion deficit argument is from Jeffrey Ding, "The diffusion deficit in scientific and technological power: re-assessing China's rise," Review of International Political Economy (2023); extended in Technology and the Rise of Great Powers: How Diffusion Shapes Economic Competition (Princeton University Press, 2024). Graduate output and publication trends are documented in bibliometric research and widely reported in the science policy press.

Small open economies navigating between US and EU frameworks while maintaining their own research policy traditions. Foundation capital at NNF scale — DKK 11.8 billion in grants and investments in 2025, with 21% now going outside Denmark³ — is geopolitically significant: one of the few sources of large-scale European research funding not subject to government priorities or electoral cycles. The Wellcome Trust in the UK is a structurally comparable model at similar scale — a century-old charitable trust built from pharmaceutical assets, professionally governed, and operating with significant independence from government priorities. How foundations respond to industrial policy pressure — by aligning, by maintaining independence, or by explicitly filling gaps left by policy — is an open and consequential question for Danish research governance.

Sources: NNF figures from 2025 Year in Review.³ The broader framing is analytical.

Denmark's political parties agreed on a four-year research and innovation framework described as the largest in Danish history. The most consequential structural shift is the earmarking of a significant majority of funding for defence and critical technology research — reflecting both NATO obligations and the post-Ukraine reassessment of strategic dependencies in energy, materials, and advanced technology. The agreement is establishing up to four university-based Defence Technology Research and Innovation Centres, a new institutional form for Danish universities with unclear long-term implications for their role as broad, civilian knowledge institutions.

The open question is whether this reorientation is a temporary response to an acute geopolitical moment or the beginning of a structural shift in what Danish universities are for. The speed of the change — negotiated and agreed within the 2025–26 political cycle — has outpaced any serious institutional debate about the consequences.

Sources: Widely reported in Danish research and political press. The institutional implications — the creation of a new dual-use research form within universities — are analytical inference from the agreement's publicly reported terms.

NIH and NSF cuts, DEI restrictions, indirect cost caps, and DOGE-driven defunding represent the most significant structural shock to the US research system in generations. The FY2027 proposed budget would cut NSF from $8.8 billion to $4 billion and NIH by 10.5% on top of earlier cuts. A STAT national survey of ~1,000 NIH-funded researchers (March 2026) documents the operational reality: a quarter have laid off lab members, 40% have cancelled planned research, two-thirds have counselled students away from academic careers, 41% have shifted research priorities to align with administration preferences, and NSF is currently awarding grants at 20% of its normal annual pace.

The European response has been larger and more coordinated than initially expected. The EU launched "Choose Europe for Science" (€500M, 2025–27) to attract displaced researchers; France announced €115M to relocate 41 research groups. ERC early-career applications from US-based researchers rose from 60 (2024 call) to 116 (2025) to 169 (2026). Several Nordic universities have dedicated recruitment initiatives. Ukrainian scientists are simultaneously being integrated into EU frameworks on expedited timelines — a second displacement dynamic European institutions are managing in parallel.

Sources: Extensively covered. STAT national survey: statnews.com, March 19, 2026. See also Stuart Buck & Aishwarya Khanduja, "A Reality-Based View of Government Funding of Science" (February 2026); Santi Ruiz, "What's Wrong with NIH Grants?" (January 2026); IFP and FAS policy outputs throughout 2025–26.⁵ ERC application figures from STAT News brain-drain reporting (December 2025).

Political, demographic, and financial pressures on universities are simultaneous across most Western countries: enrolment declines in some markets, political hostility in others, indirect cost disputes everywhere. The dominant institution for science is under strain at the same moment new alternatives are being deliberately built — which is part of what makes the current period structurally unusual.

In Denmark, a government-negotiated master's degree reform and a deliberate reduction of student places — 2,654 fewer places across universities in 2025–2029 — represent an explicit policy of constraining university throughput. This is not fiscal crisis management; it is the state actively reshaping what universities are for. Combined with the defence research reorientation in the Research and Innovation Agreement (see Category 4), it marks a significant government-directed change in the social contract of Danish higher education.

Sources: Widely reported. The simultaneous emergence of new institutional alternatives is the central argument of the diversity-problem primer, which draws on Reinhardt, Marblestone, and others.⁶ The Danish student place reduction figures are from Danish higher education ministry reporting.

The systematic restriction of academic freedom by governments hostile to universities as autonomous institutions has widened geographically. Hungary's effective dismantlement of Central European University and subordination of the Hungarian Academy of Sciences established a template. Turkey's post-2016 purge removed over 6,000 academics on national security grounds. In India, the restructuring of research funding and curriculum is increasingly constraining institutional autonomy. In Israel, legislation targeting university governance and politically inconvenient research has been advancing. Within Western Europe, the Netherlands and Germany have seen sustained political pressure — not at the same scale, but from inside liberal democratic systems, which is structurally harder to resist. The Academic Freedom Index Update 2026 (Friedrich-Alexander-Universität / V-Dem Institute, 179 countries, 2,357 expert raters) gives the US case a hard number: institutional autonomy fell from 2.4 to 1.7 (on a 0–4 scale) in a single year — the sharpest recorded annual decline — while 43 countries experienced significant deterioration over 2015–2025. The US is the most acute current instance, but part of a wider pattern rather than an exception.

Sources: Academic freedom restrictions in Hungary, Turkey, India, and Israel are documented by Scholars at Risk and widely reported. The European cases are covered in Times Higher Education and Research Professional. US and global figures from Academic Freedom Index Update 2026 (Friedrich-Alexander-Universität / V-Dem Institute, March 2026).

Gates, Bezos, Schmidt Futures, and a growing cohort of tech-wealth foundations are large, fast, and opinionated. They are reshaping what gets funded outside government channels — and developing strong views about institutional form (FROs, ARPAs, new institutes). Their priorities are not democratically determined and their influence on the direction of research is substantial and growing. Enterprise foundations like NNF — structurally different, rooted in industrial ownership rather than individual wealth — are covered under Category 4.

Sources: Nan Ransohoff, "The Third Wave of American Philanthropy" (May 2026) provides a useful framing. The general trend is widely documented.

Boom-bust cycles create pressure for faster translation from research to application; pull talent out of academia; shape what problems are considered tractable. The VC model works well for some kinds of research (drug development, software-adjacent biology) and poorly for others (fundamental physics, long-horizon materials). Its growing influence on research priorities is a structural feature of the current ecosystem, not a temporary condition.

Sources: Widely reported and discussed in the early-stage deeptech funding literature. See Aimee Rose, Andrew Chang & Raylene Yung, "The Early Stage Deep Tech Funding Gap" (March 2026).

Testing new models for IP ownership, participation, and governance — at a scale that is no longer marginal. Over $1 billion in on-chain capital has been allocated across DeSci protocols by early 2026, up from ~$60 million in combined funding in late 2024. The proof-of-concept moment: VitaDAO-funded longevity biotech Gero signed a licensing deal with Roche subsidiary Chugai in July 2025, valued at up to $250 million in milestone payments. Molecule, LabDAO, and others are extending the model beyond longevity into biotech and open science infrastructure. The model remains volatile and concentrated; the Gero/Roche deal is one data point, not a trend.

Sources: On-chain capital and ecosystem figures from World Science DAO and KuCoin DeSci sector reports (2026). VitaDAO/Gero/Chugai deal from press reporting, July 2025. See the DeSci/DAO section of the diversity-problem primer for a broader overview.⁶

Longevity, dementia, and age-related disease are becoming politically unavoidable research priorities across most rich countries. Creates strong funding gravity toward translational biomedical research and away from basic science in fields without obvious health applications. In Denmark, health remains the largest single giving category for industrial foundations at DKK 5.1 billion in 2024.⁴

Sources: The general demographic trend is widely documented. The DKK 5.1bn health figure is from Fondenes Videnscenter.⁴

Research output from the Global South is rising substantially, but the pattern is not uniform. China dominates the bibliometric growth; India and Southeast Asia are expanding significantly. Much of sub-Saharan Africa is not meaningfully participating in this shift — research capacity there remains thin, unevenly distributed, and heavily donor-dependent. The aggregate trend changes where knowledge is produced and for whom, and creates pressure on funding models and collaboration structures built around Northern institutions as the default producers and gatekeepers. But describing this as a uniform "Global South rise" obscures the widening internal gap between an emerging scientific China-India axis and large parts of the world that remain effectively excluded.

Sources: The trend is well documented in bibliometric research. The NNF's geographic expansion — with explicit focus on East Africa, South and Southeast Asia³ — reflects a specific institutional response to the unevenness of this shift.

Growing demand for translational psychiatry and public mental health research; current institutions are poorly positioned to deliver it at scale. Also intersects directly with the academic labour force question: the mental health consequences of precarious academic careers are increasingly documented and publicly debated, particularly in the Nordic countries.

Sources: The general crisis is widely reported. Stuart Buck's analysis of NIMH under Tom Insel — "Mental Health at NIH" (May 2026) — documents the institutional failure to translate $20 billion in investment into improved patient outcomes. The Danish academic workplace dimension is discussed in category 9.

Is actively compromising the integrity of the scientific record and creating new demands on verification infrastructure that most institutions have not yet built. Fabricated references have risen 12-fold since 2023: a Retraction Watch analysis (May 2026) finds 1 in 277 PubMed-indexed papers now contains fabricated citations, up from 1 in 458 in 2025 and 1 in 2,828 in 2023. 100 hallucinated citations appeared in accepted NeurIPS 2025 papers; arXiv introduced a 1-year submission ban for incontrovertible LLM-generated errors in 2026. Peer review systems designed for a world of scarce submissions are poorly equipped for a world of abundant AI-assisted manuscripts. The institutions that maintain the scientific record — journals, preprint servers, data repositories — need structural redesign.

Sources: Fabricated-reference rate from Retraction Watch analysis, May 7, 2026; arXiv LLM error policy (2026). See also Snezana Gvozdenovic, "The Burden of Discernment" (Palladium, May 2026) on the structural argument about generative AI and institutional capacity for knowledge discernment.

Public trust in scientific institutions has eroded in most Western countries — partly due to high-profile failures of reproducibility, partly due to politicisation of scientific questions during and after COVID-19. One structural cause: major research programmes fail routinely, but insiders rarely discuss this publicly for fear of professional retaliation, while less-informed outside voices fill the vacuum — a market-for-lemons dynamic that impairs the field's capacity to self-correct.⁷

Sources: Stuart Buck & Ben Reinhardt, "The Paradox of Harmful Information", Arena Magazine (October 2025).⁷

Open data, open code, and preprints are shifting from advocacy positions to expected norms — at different speeds across fields. Changes who can access and build on research; reduces some barriers to participation; creates new questions about credit, quality control, and the economics of scientific publishing.

Sources: The shift to open norms is well documented in science policy literature. See RoRI Working Paper No. 20 (May 2026) for context on how publishing norms intersect with funding structures. For the foundational theoretical argument that open science and proprietary R&D represent structurally incompatible institutional logics — and that universities navigating both simultaneously face an irresolvable tension — see Paul A. David (posthumous, ed. Bronwyn H. Hall & Jacques Mairesse), "Research universities' futures in the networked world," Economics of Innovation and New Technology (June 2026).

The infrastructure through which science reaches public understanding has been substantially degraded over the past two decades. Science journalism has contracted sharply: specialist desks have closed across major newspapers; the editors and reporters capable of translating technical findings accurately have largely moved into institutional communications roles, where their job is advocacy rather than explanation. Social media fragmented the broadcast-era channels that allowed scientific findings to reach large audiences through shared intermediaries. The more consequential effect is field-specific: certain domains — vaccine safety, gender and sex research, race and genetics, climate attribution — have become politically marked in ways that constrain what universities can visibly fund, what researchers can publish without reputational risk, and what politicians will publicly defend. This is not primarily a misinformation problem; it is a structural problem about which epistemic authorities are trusted, by whom, and on which questions. Pew Research (January 2026, n=5,111) puts a number on the divergence: overall trust in science holds at 77%, but Democrats are 28 percentage points more likely than in 2023 to say the US is losing ground in science internationally, while Republican confidence in scientists has declined sharply — aggregate stability concealing deepening asymmetric erosion. It shapes the political feasibility of research funding and the behaviour of researchers in ways that don't appear in any formal governance document.

Sources: The contraction of science journalism is widely documented in media industry research. The field-specific chilling effects are increasingly discussed in academic freedom and research integrity literature. The structural argument — about trusted authorities rather than misinformation — is analytical. Trust figures from Pew Research Center, Trust in Science Survey (January 15, 2026).

The university career path — long chains of fixed-term contracts, postdoc to postdoc, with permanent positions scarce and delayed by a decade or more after the PhD — selects against people with families, outside options, or low risk tolerance. The people who remain are not a neutral sample of those who could do the work. This is a structural feature of the system, not an incidental problem, and it shapes what science gets done and who does it.

Sources: The precarity pattern is widely documented in academic labour literature. Prachee Avasthi, "Free the PhD" (February 2026) addresses the career structure directly. The selection-effects argument is analytical.

How researchers are formally evaluated shapes what they optimise for more directly than most external forces in this document. The journal impact factor — the dominant metric of academic productivity for three decades — is now being actively dismantled as a primary assessment tool. The Coalition for Advancing Research Assessment (CoARA) — 700+ signatory institutions, National Chapters grown from 5 (2023) to 18 (2025), 25 cascade-funded reform projects running — represents commitments from 40+ countries to move toward broader, contextual evaluation. In the UK, the Research Excellence Framework (REF) directly shapes institutional strategy and resource allocation at a structural level. The transition is consequential: impact factor optimisation pushed researchers toward incremental results in high-volume journals and created the conditions for the reproducibility failures documented in Category 8. What replaces it will shape the next generation's research choices in ways not yet visible. The reform is slower than the rhetoric suggests — CoARA commitments are non-binding and institutional cultures change slowly — but the direction is set, and the researchers and institutions that adapt earliest will have a genuine competitive advantage.

Sources: CoARA signatories and commitments are documented at coara.eu. REF methodology is publicly documented by Research England. The behavioural consequences of metric-driven assessment are extensively discussed in the research evaluation literature. The connection to reproducibility failures is analytical inference from the wider research integrity literature.

Sustained public focus in Denmark on poor working conditions, inadequate management practices, and the mental health consequences of academic employment. Universities are increasingly perceived as organisations that extract rather than develop talent — demanding high output under precarious conditions with limited support structures. This is not unique to Denmark, but the domestic debate has been notably sharp and is increasingly connected to questions of institutional reform.

In 2026, Forsker-bevægelsen (The Researchers' Movement) published 130 anonymised testimonies from researchers at Danish universities, collected through a public #SafeToSpeak campaign running August–October 2025. The accounts document a consistent pattern across institutions: retaliation for speaking up — including threats of dismissal, non-renewal of contracts, and exclusion from professional networks; management cultures in which fear of consequences suppresses both internal criticism and public commentary; authorship and research-integrity disputes handled through institutional pressure rather than formal integrity processes; and PhD students facing acute vulnerability when supervisors control both academic progress and continued employment. Several testimonies describe deliberate self-censorship on research topics, public debate, and workplace assessment surveys. The document is self-selected and does not claim to be statistically representative, but the volume, institutional spread, and specificity of the accounts indicate structural rather than idiosyncratic problems.

Sources: Forsker-bevægelsen, 130 Testimonies from the #SafeToSpeak Campaign (August–October 2025, published 2026). The broader Danish debate has been covered in Magisterbladet, Politiken, and academic union publications.

Tech, biotech, and consulting offer better pay, clearer career progression, and more autonomy than academic positions. The talent pool available to academic science is being skimmed before it reaches permanent positions. The researchers most likely to thrive in industry — those with outside options, risk tolerance, and strong technical skills — are exactly those most likely to leave academia. This creates a slow adverse selection dynamic that compounds over time.

Sources: The talent competition pattern is widely reported. The adverse selection argument is analytical inference from the general precarity and compensation literature.

FROs, enterprise-foundation-funded institutes, and ARPA-model programmes are partly attractive because they offer what universities have stopped offering: stable employment, defined problems, autonomy within structure, and an exit that doesn't feel like failure. Convergent Research and ARIA launched two new UK FROs — Meridial and Echo Labs — in April 2026, the first to emerge from Convergent's residency programme as an ARIA Activation Partner. The new institutions conversation is not only about doing different science — it is about offering a different relationship between a researcher and their work.⁸ Whether these models can absorb enough of the talent pipeline to matter at scale is an open question.

Sources: The labour-reform framing is developed in the diversity-problem primer, drawing on Marblestone & Rodriques (2020) and Reinhardt.⁸