Global Science and Engineering Frontier Risk and Resilience Report 2026: Baseline Edition, The State of the World’s Science and Engineering Frontier.

S.1Abstract

S.2Key messages

Six readings that frame the science and engineering baseline, and the eight questions this edition answers, each with its status and source.

1. About USD 3.48 trillion in 2024; the share of GDP depends on the issuer.¹²³

   How much does the world spend on research?Reported

2. China and the United States together perform more than half of spending; researcher density differs widely across regions.¹²

   Who does the research?Reported

3. Yes. 3.5 million articles in 2024 and 284,486 arXiv submissions in 2025.¹⁵

   Is the literature growing?Reported

4. Retractions and paper-mill fraud are rising pressures; the share of the record affected is not known.⁶¹⁴

   Can the record be trusted?Reported

5. It is a growing part of output, but agents reach about half of expert scores on research tasks.⁹

   What is AI doing in science?Reported

6. Five exascale systems exist on the TOP500 list; industry builds most notable models.⁴⁸

   Who has frontier compute?Reported

7. Large open archives exist for structures and species records.¹²²²²³

   Is biology data open?Reported

8. A loss of trust in the research record during an emergency. SCN-GS15-INT-01

   What will 2027 test first?Expert input

S.3Headline finding

ReportedEQL 5NRF-SECTOR-SCIENCE-GLOBAL-2026-BASELINE-DAT-003

In 2023, the world had 1,486 researchers per million people, with 88 in Sub-Saharan Africa and 5,006 in Northern America.

The distribution of people who can do research is far more uneven than the distribution of articles or spending. Least Developed Countries stand at 70 researchers per million and spend 0.28% of GDP on research, so knowledge made elsewhere is the main input to their decisions². Women were 31.4% of researchers worldwide, with regional shares from 23.9% in Eastern Asia to 45.0% in Northern Africa².

S.4At a glance

S.5What each steward reads in it

S.6What this edition is, and is not

Notes

Numbers are the edition’s reference numbers, so a note reads the same in every chapter. The full list, with every source and its licence, is in Annex C. Annex C, references

1. 1NSB (ed.), Science and Engineering Indicators 2026: The State of U.S. Science and Engineering (NSB-SEP-2026-1), May 2026, National Science Board and National Center for Science and Engineering Statistics, US National Science Foundation, Alexandria. Full PDF opened; global R&D, publications, collaboration, venture investment, semiconductor and doctoral-degree figures read from the text. Observation year 2024. Accessed 2026-11-03.back
2. 2UNESCO Institute for Statistics (ed.), UIS data API, data version 20260507 (February 2026 data release): indicators EXPGDP.TOT, RESDEN.INHAB.TFTE and FRESP.THC, world and regional aggregates. UNESCO, Montreal. Data pulled and read. Observation year 2023. Accessed 2026-11-03.back
3. 5POST-CLOSE, context only: arXiv (ed.), Monthly submissions statistics (monthly CSV, 1991 to November 2026; annual totals summed by the editors). Cornell University, Ithaca. Observation year 2025. Accessed 2026-11-03.back
4. 6Crossref (ed.), Crossref REST API, work counts by type and publication year and counts of retraction notices (update-type filter), queried live on the access date. Crossref, Oxford. Counts are subject to later registration. Observation year 2025 (series 2015 to 2025). Accessed 2026-11-03.back
5. 14Richardson, R.A.K. et al., The entities enabling scientific fraud at scale are large, resilient, and growing rapidly, Proceedings of the National Academy of Sciences 122(32), 2025 (abstract and open full text record read through Europe PMC). National Academy of Sciences, Washington. Observation year 2025. Accessed 2026-11-03.back
6. 9Stanford HAI (ed.), The 2026 AI Index Report, Chapter 5 Science, chapter page with its highlights. Stanford Institute for Human-Centered AI, Stanford. Page read; the chapter PDF was not opened. Observation year 2025. Accessed 2026-11-03.back
7. 4TOP500 project (ed.), TOP500 list, 67th edition, June 2026 (release text read in full for the leading systems). TOP500, Prometeus GmbH. Observation year 2026. Accessed 2026-11-03.back
8. 10Google Quantum AI and collaborators, Quantum error correction below the surface code threshold, arXiv 2408.13687 (full text opened) and Nature 638, 920 to 926, 2024. Observation year 2024. Accessed 2026-11-03.back
9. 3DATA VINTAGE AFTER CLOSE: World Bank (ed.), World Development Indicators API (series updated 2026-10-08, after the 30 September evidence close; cited as the position at access): GB.XPD.RSDV.GD.ZS, SP.POP.SCIE.RD.P6, IP.JRN.ARTC.SC, IP.PAT.RESD, IP.PAT.NRES, world values. World Bank, Washington. Data pulled and read. Observation year 2023 (patents 2024, researchers 2020). Accessed 2026-11-03.back
10. 8Stanford HAI (ed.), The 2026 AI Index Report, overview page with the ten top takeaways. Stanford Institute for Human-Centered AI, Stanford. Page read; the report PDF was not opened. Observation year 2025. Accessed 2026-11-03.back
11. 12Varadi, M. et al., AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences, Nucleic Acids Research 52 (database issue), 368 to 375, 2024 (abstract read through Europe PMC). Oxford University Press, Oxford. Observation year 2024. Accessed 2026-11-03.back
12. 22RCSB Protein Data Bank (ed.), Current holdings, list of entry identifiers (262,922 entries counted by the editors on the access date). Research Collaboratory for Structural Bioinformatics. Observation year 2026. Accessed 2026-11-03.back
13. 23GBIF (ed.), GBIF API, occurrence count and counts by year (3,890,294,224 records on the access date; 305,893,347 dated 2025). Global Biodiversity Information Facility, Copenhagen. Observation year 2026. Accessed 2026-11-03.back

1.1The questions this part answers

Key questions

• Where is science and engineering risk concentrated?
• Which couplings carry science and engineering shocks into other systems?
• Which science and engineering capabilities hold under compound stress?
• What must 2027 evidence measure first?

In this part

• Narrative
• Headline finding
• Lead figure
• Five coupling chapters
• Twelve-domain reading
• Sector screens
• Dependency matrix
• Risk objects
• Capability ladder
• Scenario
• Three lenses
• Portfolio

1.2The argument

1.3Twenty couplings between five systems

The science and engineering-centred matrix types the twenty couplings between the five core systems, read for their science and engineering content. Strengths are Nexus estimates (ESTIMATED) from the cited evidence and analogue events; they set the graph that 2027 evidence will test.

Transmission paths

1. 01

   BIO → HLT

   Sequence, structure and trial data feed medicines and gene therapy.

   Reported AlphaFold database and the 2023 authorisation¹²¹⁷

2. 02

   AIC → DAT → GOV

   Mass-produced or AI-assisted papers dilute the record and strain peer review.

   Reported paper-mill case studies¹⁴¹

3. 03

   FIN → DAT

   Funding and talent concentrate capacity in a few regions.

   Reported R&D shares and researcher density¹²

4. 04

   INF → AIC

   Exascale and training compute decide which groups can work at the frontier.

   Reported TOP500 list and compute trend⁴¹³

5. 05

   TRD → DAT

   Co-authorship and mobility carry knowledge across borders.

   Reported 22% of articles; AI researcher moves¹⁸

1.4Reinforcing loops

A shock entering either system of a loop can return amplified through the other.

Notes

Numbers are the edition’s reference numbers, so a note reads the same in every chapter. The full list, with every source and its licence, is in Annex C. Annex C, references

1. 11Abramson, J. et al., Accurate structure prediction of biomolecular interactions with AlphaFold 3, Nature 630, 493 to 500, 2024 (abstract and open full text read through Europe PMC). Springer Nature, London. Observation year 2024. Accessed 2026-11-03.back
2. 17US Food and Drug Administration (ed.), FDA Approves First Gene Therapies to Treat Patients with Sickle Cell Disease, press announcement of 8 December 2023. FDA, Silver Spring. Page read. Observation year 2023. Accessed 2026-11-03.back
3. 9Stanford HAI (ed.), The 2026 AI Index Report, Chapter 5 Science, chapter page with its highlights. Stanford Institute for Human-Centered AI, Stanford. Page read; the chapter PDF was not opened. Observation year 2025. Accessed 2026-11-03.back
4. 4TOP500 project (ed.), TOP500 list, 67th edition, June 2026 (release text read in full for the leading systems). TOP500, Prometeus GmbH. Observation year 2026. Accessed 2026-11-03.back
5. 1NSB (ed.), Science and Engineering Indicators 2026: The State of U.S. Science and Engineering (NSB-SEP-2026-1), May 2026, National Science Board and National Center for Science and Engineering Statistics, US National Science Foundation, Alexandria. Full PDF opened; global R&D, publications, collaboration, venture investment, semiconductor and doctoral-degree figures read from the text. Observation year 2024. Accessed 2026-11-03.back
6. 2UNESCO Institute for Statistics (ed.), UIS data API, data version 20260507 (February 2026 data release): indicators EXPGDP.TOT, RESDEN.INHAB.TFTE and FRESP.THC, world and regional aggregates. UNESCO, Montreal. Data pulled and read. Observation year 2023. Accessed 2026-11-03.back
7. 5POST-CLOSE, context only: arXiv (ed.), Monthly submissions statistics (monthly CSV, 1991 to November 2026; annual totals summed by the editors). Cornell University, Ithaca. Observation year 2025. Accessed 2026-11-03.back
8. 6Crossref (ed.), Crossref REST API, work counts by type and publication year and counts of retraction notices (update-type filter), queried live on the access date. Crossref, Oxford. Counts are subject to later registration. Observation year 2025 (series 2015 to 2025). Accessed 2026-11-03.back
9. 14Richardson, R.A.K. et al., The entities enabling scientific fraud at scale are large, resilient, and growing rapidly, Proceedings of the National Academy of Sciences 122(32), 2025 (abstract and open full text record read through Europe PMC). National Academy of Sciences, Washington. Observation year 2025. Accessed 2026-11-03.back
10. 16Open Science Collaboration, Estimating the reproducibility of psychological science, Science 349(6251), aac4716, 2015 (abstract read through Europe PMC; full text not opened). American Association for the Advancement of Science, Washington. Observation year 2015. Accessed 2026-11-03.back
11. 13Sevilla, J. et al., Compute Trends Across Three Eras of Machine Learning, arXiv 2202.05924, 2022 (abstract page read). Cornell University, Ithaca. Observation year 2022. Accessed 2026-11-03.back
12. 27Gottweis, J. et al., Accelerating scientific discovery with Co-Scientist, arXiv 2502.18864 (version 2, 2026; abstract read through the arXiv API). Cornell University, Ithaca. Observation year 2025. Accessed 2026-11-03.back
13. 25UNESCO (ed.), UNESCO Recommendation on Open Science, 2021, programme page (193 Member States). UNESCO, Paris. Page read; the Recommendation text itself returned an access error. Observation year 2021. Accessed 2026-11-03.back
14. 26UNESCO (ed.), Recommendation on the Ethics of Artificial Intelligence, November 2021, programme page (193 Member States). UNESCO, Paris. Page read; the text itself was not opened. Observation year 2021. Accessed 2026-11-03.back
15. 18WHO (ed.), Global guidance framework for the responsible use of the life sciences: mitigating biorisks and governing dual-use research, 174 pages, 2022. World Health Organization, Geneva. Publication page and overview read; the full text was not opened. Observation year 2022. Accessed 2026-11-03.back
16. 12Varadi, M. et al., AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences, Nucleic Acids Research 52 (database issue), 368 to 375, 2024 (abstract read through Europe PMC). Oxford University Press, Oxford. Observation year 2024. Accessed 2026-11-03.back
17. 8Stanford HAI (ed.), The 2026 AI Index Report, overview page with the ten top takeaways. Stanford Institute for Human-Centered AI, Stanford. Page read; the report PDF was not opened. Observation year 2025. Accessed 2026-11-03.back

2.1Headline finding

ReportedEQL 5NRF-SECTOR-SCIENCE-GLOBAL-2026-BASELINE-DAT-003

In 2023, the world had 1,486 researchers per million people, with 88 in Sub-Saharan Africa and 5,006 in Northern America.

Implication

The distribution of people who can do research is far more uneven than the distribution of articles or spending. Least Developed Countries stand at 70 researchers per million and spend 0.28% of GDP on research, so knowledge made elsewhere is the main input to their decisions². Women were 31.4% of researchers worldwide, with regional shares from 23.9% in Eastern Asia to 45.0% in Northern Africa².

What would move it

The next UNESCO data release, national statistics on researcher headcount and the reconciliation with the World Bank series would each move this finding.

2.2Finance

1. ReportedEQL 4NRF-SECTOR-SCIENCE-GLOBAL-2026-BASELINE-FIN-001

   Worldwide R&D expenditure totalled USD 3.48 trillion in 2024 on a comparable basis; China held 30% and the United States 29%, and the EU-27 18%.¹

2. ReportedEQL 4NRF-SECTOR-SCIENCE-GLOBAL-2026-BASELINE-FIN-002

   World R&D intensity is 1.92% of GDP in 2023 in the UNESCO series, up from 1.71% in 2015, while the World Bank series shows 2.60% for 2023; the two issuers are not reconciled here.²³

2.3Data and digital

1. ReportedEQL 5NRF-SECTOR-SCIENCE-GLOBAL-2026-BASELINE-DAT-003

   The world had 1,486 researchers per million people in 2023, against 1,141 in 2015, with 88 in Sub-Saharan Africa, 347 in Southern Asia and 5,006 in Northern America; women were 31.4% of researchers.²

2.4Trade

1. ReportedEQL 4NRF-SECTOR-SCIENCE-GLOBAL-2026-BASELINE-TRD-004

   Worldwide scientific output was 3.5 million articles in 2024, led by China (31%), the United States (12%) and India (7%), and 22% of articles had international co-authors, up from 19% in 2012.¹

2.5Governance

1. ReportedEQL 3NRF-SECTOR-SCIENCE-GLOBAL-2026-BASELINE-GOV-005

   Crossref registered 15,582 retraction notices in 2023, 9,700 in 2024 and 7,632 in 2025, against 1,678 in 2015, and PNAS case studies conclude that fraudulent publications grow faster than legitimate science.⁶¹⁴¹

2.6AIC

1. ReportedEQL 3NRF-SECTOR-SCIENCE-GLOBAL-2026-BASELINE-AIC-006

   AI accounts for 5.8% to 8.8% of scientific research output depending on the field, up from below 1% in 2010, and the best AI agents score 38.8% against 83.5% for PhD experts on a research-task benchmark.⁹

2. ReportedEQL 4NRF-SECTOR-SCIENCE-GLOBAL-2026-BASELINE-AIC-007

   Training compute of notable ML systems doubled about every six months since the early 2010s, against about twenty months before 2010, and industry produced more than 90% of notable frontier models in 2025.¹³⁸

2.7Infrastructure

1. ReportedEQL 5NRF-SECTOR-SCIENCE-GLOBAL-2026-BASELINE-INF-008

   The June 2026 TOP500 list holds five exascale systems; LineShine in Shenzhen leads with 2.198 Eflop/s, and JUPITER Booster is the first European exascale system.⁴

2. ReportedEQL 4NRF-SECTOR-SCIENCE-GLOBAL-2026-BASELINE-INF-009

   A 101-qubit distance-7 surface-code memory showed logical error suppressed by a factor of 2.14 for each step in code distance, with 0.143% error per cycle.¹⁰

2.8Biodiversity

1. ReportedEQL 4NRF-SECTOR-SCIENCE-GLOBAL-2026-BASELINE-BIO-010

   The AlphaFold database holds over 214 million predicted protein structures, the Protein Data Bank held 262,922 entries and GBIF held 3.89 billion occurrence records at access.¹²²²²³

2.9Health

1. ReportedEQL 4NRF-SECTOR-SCIENCE-GLOBAL-2026-BASELINE-HLT-011

   The FDA gave marketing authorisation on 8 December 2023 to the first treatment using a novel genome-editing technology, and WHO published a framework in 2022 on governing dual-use life science research.¹⁷¹⁸

2.10Systemic

1. Expert inputEQL 2NRF-SECTOR-SCIENCE-GLOBAL-2026-BASELINE-SYS-012

   The leading compound pathway for 2027 testing is a loss of trust in the research record during a fast-moving emergency. § scenario

Notes

Numbers are the edition’s reference numbers, so a note reads the same in every chapter. The full list, with every source and its licence, is in Annex C. Annex C, references

1. 2UNESCO Institute for Statistics (ed.), UIS data API, data version 20260507 (February 2026 data release): indicators EXPGDP.TOT, RESDEN.INHAB.TFTE and FRESP.THC, world and regional aggregates. UNESCO, Montreal. Data pulled and read. Observation year 2023. Accessed 2026-11-03.back
2. 1NSB (ed.), Science and Engineering Indicators 2026: The State of U.S. Science and Engineering (NSB-SEP-2026-1), May 2026, National Science Board and National Center for Science and Engineering Statistics, US National Science Foundation, Alexandria. Full PDF opened; global R&D, publications, collaboration, venture investment, semiconductor and doctoral-degree figures read from the text. Observation year 2024. Accessed 2026-11-03.back
3. 3DATA VINTAGE AFTER CLOSE: World Bank (ed.), World Development Indicators API (series updated 2026-10-08, after the 30 September evidence close; cited as the position at access): GB.XPD.RSDV.GD.ZS, SP.POP.SCIE.RD.P6, IP.JRN.ARTC.SC, IP.PAT.RESD, IP.PAT.NRES, world values. World Bank, Washington. Data pulled and read. Observation year 2023 (patents 2024, researchers 2020). Accessed 2026-11-03.back
4. 6Crossref (ed.), Crossref REST API, work counts by type and publication year and counts of retraction notices (update-type filter), queried live on the access date. Crossref, Oxford. Counts are subject to later registration. Observation year 2025 (series 2015 to 2025). Accessed 2026-11-03.back
5. 14Richardson, R.A.K. et al., The entities enabling scientific fraud at scale are large, resilient, and growing rapidly, Proceedings of the National Academy of Sciences 122(32), 2025 (abstract and open full text record read through Europe PMC). National Academy of Sciences, Washington. Observation year 2025. Accessed 2026-11-03.back
6. 9Stanford HAI (ed.), The 2026 AI Index Report, Chapter 5 Science, chapter page with its highlights. Stanford Institute for Human-Centered AI, Stanford. Page read; the chapter PDF was not opened. Observation year 2025. Accessed 2026-11-03.back
7. 13Sevilla, J. et al., Compute Trends Across Three Eras of Machine Learning, arXiv 2202.05924, 2022 (abstract page read). Cornell University, Ithaca. Observation year 2022. Accessed 2026-11-03.back
8. 8Stanford HAI (ed.), The 2026 AI Index Report, overview page with the ten top takeaways. Stanford Institute for Human-Centered AI, Stanford. Page read; the report PDF was not opened. Observation year 2025. Accessed 2026-11-03.back
9. 4TOP500 project (ed.), TOP500 list, 67th edition, June 2026 (release text read in full for the leading systems). TOP500, Prometeus GmbH. Observation year 2026. Accessed 2026-11-03.back
10. 10Google Quantum AI and collaborators, Quantum error correction below the surface code threshold, arXiv 2408.13687 (full text opened) and Nature 638, 920 to 926, 2024. Observation year 2024. Accessed 2026-11-03.back
11. 12Varadi, M. et al., AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences, Nucleic Acids Research 52 (database issue), 368 to 375, 2024 (abstract read through Europe PMC). Oxford University Press, Oxford. Observation year 2024. Accessed 2026-11-03.back
12. 22RCSB Protein Data Bank (ed.), Current holdings, list of entry identifiers (262,922 entries counted by the editors on the access date). Research Collaboratory for Structural Bioinformatics. Observation year 2026. Accessed 2026-11-03.back
13. 23GBIF (ed.), GBIF API, occurrence count and counts by year (3,890,294,224 records on the access date; 305,893,347 dated 2025). Global Biodiversity Information Facility, Copenhagen. Observation year 2026. Accessed 2026-11-03.back
14. 17US Food and Drug Administration (ed.), FDA Approves First Gene Therapies to Treat Patients with Sickle Cell Disease, press announcement of 8 December 2023. FDA, Silver Spring. Page read. Observation year 2023. Accessed 2026-11-03.back
15. 18WHO (ed.), Global guidance framework for the responsible use of the life sciences: mitigating biorisks and governing dual-use research, 174 pages, 2022. World Health Organization, Geneva. Publication page and overview read; the full text was not opened. Observation year 2022. Accessed 2026-11-03.back

3.1Where each system stood at evidence close

3.2Domain posture

Each of the twelve coupled domains is read through science and engineering, in words, not scored. The coupling column names the dependency that carries science and engineering risk into each domain.

3.3Sector screens

All sixteen sectors are screened for their dependence on science and engineering, on the same grammar as every national edition. The evidence column describes coverage of the baseline evidence, not the performance of the sector.

Notes

Numbers are the edition’s reference numbers, so a note reads the same in every chapter. The full list, with every source and its licence, is in Annex C. Annex C, references

1. 1NSB (ed.), Science and Engineering Indicators 2026: The State of U.S. Science and Engineering (NSB-SEP-2026-1), May 2026, National Science Board and National Center for Science and Engineering Statistics, US National Science Foundation, Alexandria. Full PDF opened; global R&D, publications, collaboration, venture investment, semiconductor and doctoral-degree figures read from the text. Observation year 2024. Accessed 2026-11-03.back
2. 2UNESCO Institute for Statistics (ed.), UIS data API, data version 20260507 (February 2026 data release): indicators EXPGDP.TOT, RESDEN.INHAB.TFTE and FRESP.THC, world and regional aggregates. UNESCO, Montreal. Data pulled and read. Observation year 2023. Accessed 2026-11-03.back
3. 4TOP500 project (ed.), TOP500 list, 67th edition, June 2026 (release text read in full for the leading systems). TOP500, Prometeus GmbH. Observation year 2026. Accessed 2026-11-03.back
4. 6Crossref (ed.), Crossref REST API, work counts by type and publication year and counts of retraction notices (update-type filter), queried live on the access date. Crossref, Oxford. Counts are subject to later registration. Observation year 2025 (series 2015 to 2025). Accessed 2026-11-03.back
5. 9Stanford HAI (ed.), The 2026 AI Index Report, Chapter 5 Science, chapter page with its highlights. Stanford Institute for Human-Centered AI, Stanford. Page read; the chapter PDF was not opened. Observation year 2025. Accessed 2026-11-03.back
6. 17US Food and Drug Administration (ed.), FDA Approves First Gene Therapies to Treat Patients with Sickle Cell Disease, press announcement of 8 December 2023. FDA, Silver Spring. Page read. Observation year 2023. Accessed 2026-11-03.back
7. 11Abramson, J. et al., Accurate structure prediction of biomolecular interactions with AlphaFold 3, Nature 630, 493 to 500, 2024 (abstract and open full text read through Europe PMC). Springer Nature, London. Observation year 2024. Accessed 2026-11-03.back
8. 19CERN (ed.), Future Circular Collider, feasibility study page (results published in spring 2025: 91 km ring, estimated cost CHF 15 billion). CERN, Geneva. Page read. Observation year 2025. Accessed 2026-11-03.back
9. 23GBIF (ed.), GBIF API, occurrence count and counts by year (3,890,294,224 records on the access date; 305,893,347 dated 2025). Global Biodiversity Information Facility, Copenhagen. Observation year 2026. Accessed 2026-11-03.back
10. 25UNESCO (ed.), UNESCO Recommendation on Open Science, 2021, programme page (193 Member States). UNESCO, Paris. Page read; the Recommendation text itself returned an access error. Observation year 2021. Accessed 2026-11-03.back
11. 18WHO (ed.), Global guidance framework for the responsible use of the life sciences: mitigating biorisks and governing dual-use research, 174 pages, 2022. World Health Organization, Geneva. Publication page and overview read; the full text was not opened. Observation year 2022. Accessed 2026-11-03.back

4.1Risk objects

Each risk is read across the chain from hazard to consequence, with its status and evidence-quality level.

1. InferredEQL 3RSK-GS15-GOV-001

   Loss of trust in the research record

   Driver

   paper mills, brokers and AI-assisted text growing faster than screening¹⁴⁹

   Exposure

   clinical guidance, policy advice and public debate that cite the literature

   Vulnerability

   thin screening at some journals; replication rarely funded¹⁶

   Capacity

   retraction notices; open data; replication projects; integrity units

   Cascade

   DAT → HLT, GOV

   Horizon

   1 to 5 years

2. InferredEQL 3RSK-GS15-DAT-002

   Concentration of research capacity and compute

   Driver

   concentrated funding, people and compute¹²⁴

   Exposure

   regions with few researchers; groups without access to large machines

   Vulnerability

   dependence on foreign results and foreign tools

   Capacity

   open data; shared facilities; regional networks²⁴

   Cascade

   DAT → HLT, FOO, FIN

   Horizon

   1 to 10 years

3. InferredEQL 3RSK-GS15-BIO-003

   Dual-use risk in life science research

   Driver

   cheaper tools and models for biology¹¹¹⁸

   Exposure

   laboratories, funders, publishers and model developers

   Vulnerability

   guidance that is voluntary in many places

   Capacity

   WHO framework; national oversight; institutional review¹⁸

   Cascade

   BIO → HLT, GOV

   Horizon

   1 to 5 years

4. InferredEQL 3RSK-GS15-INF-004

   Long-lead facilities and single-funder dependence

   Driver

   multi-billion projects with decade-long schedules¹⁹

   Exposure

   accelerators, observatories, supercomputers and archives

   Vulnerability

   single sites, single funders and shifting budgets

   Capacity

   international councils; staged decisions; shared archives²²

   Cascade

   INF → DAT, FIN

   Horizon

   5 to 10 years

4.2Capability ladder

Existence is not readiness. Each of the fifteen capability functions is placed on the ladder by the strongest public evidence for science and engineering at the global level.

Notes

Numbers are the edition’s reference numbers, so a note reads the same in every chapter. The full list, with every source and its licence, is in Annex C. Annex C, references

1. 14Richardson, R.A.K. et al., The entities enabling scientific fraud at scale are large, resilient, and growing rapidly, Proceedings of the National Academy of Sciences 122(32), 2025 (abstract and open full text record read through Europe PMC). National Academy of Sciences, Washington. Observation year 2025. Accessed 2026-11-03.back
2. 9Stanford HAI (ed.), The 2026 AI Index Report, Chapter 5 Science, chapter page with its highlights. Stanford Institute for Human-Centered AI, Stanford. Page read; the chapter PDF was not opened. Observation year 2025. Accessed 2026-11-03.back
3. 16Open Science Collaboration, Estimating the reproducibility of psychological science, Science 349(6251), aac4716, 2015 (abstract read through Europe PMC; full text not opened). American Association for the Advancement of Science, Washington. Observation year 2015. Accessed 2026-11-03.back
4. 1NSB (ed.), Science and Engineering Indicators 2026: The State of U.S. Science and Engineering (NSB-SEP-2026-1), May 2026, National Science Board and National Center for Science and Engineering Statistics, US National Science Foundation, Alexandria. Full PDF opened; global R&D, publications, collaboration, venture investment, semiconductor and doctoral-degree figures read from the text. Observation year 2024. Accessed 2026-11-03.back
5. 2UNESCO Institute for Statistics (ed.), UIS data API, data version 20260507 (February 2026 data release): indicators EXPGDP.TOT, RESDEN.INHAB.TFTE and FRESP.THC, world and regional aggregates. UNESCO, Montreal. Data pulled and read. Observation year 2023. Accessed 2026-11-03.back
6. 4TOP500 project (ed.), TOP500 list, 67th edition, June 2026 (release text read in full for the leading systems). TOP500, Prometeus GmbH. Observation year 2026. Accessed 2026-11-03.back
7. 24European Commission (ed.), Horizon Europe, programme page (indicative funding of EUR 93.5 billion for 2021 to 2027 after the mid-term review). European Commission, Brussels. Page read. Observation year 2026. Accessed 2026-11-03.back
8. 11Abramson, J. et al., Accurate structure prediction of biomolecular interactions with AlphaFold 3, Nature 630, 493 to 500, 2024 (abstract and open full text read through Europe PMC). Springer Nature, London. Observation year 2024. Accessed 2026-11-03.back
9. 18WHO (ed.), Global guidance framework for the responsible use of the life sciences: mitigating biorisks and governing dual-use research, 174 pages, 2022. World Health Organization, Geneva. Publication page and overview read; the full text was not opened. Observation year 2022. Accessed 2026-11-03.back
10. 19CERN (ed.), Future Circular Collider, feasibility study page (results published in spring 2025: 91 km ring, estimated cost CHF 15 billion). CERN, Geneva. Page read. Observation year 2025. Accessed 2026-11-03.back
11. 22RCSB Protein Data Bank (ed.), Current holdings, list of entry identifiers (262,922 entries counted by the editors on the access date). Research Collaboratory for Structural Bioinformatics. Observation year 2026. Accessed 2026-11-03.back
12. 25UNESCO (ed.), UNESCO Recommendation on Open Science, 2021, programme page (193 Member States). UNESCO, Paris. Page read; the Recommendation text itself returned an access error. Observation year 2021. Accessed 2026-11-03.back
13. 37UNVERIFIED, not opened: UNESCO (ed.), Recommendation on Science and Scientific Researchers, 2017, cited from memory. UNESCO, Paris. Observation year 2017. Accessed 2026-11-03.back
14. 6Crossref (ed.), Crossref REST API, work counts by type and publication year and counts of retraction notices (update-type filter), queried live on the access date. Crossref, Oxford. Counts are subject to later registration. Observation year 2025 (series 2015 to 2025). Accessed 2026-11-03.back
15. 23GBIF (ed.), GBIF API, occurrence count and counts by year (3,890,294,224 records on the access date; 305,893,347 dated 2025). Global Biodiversity Information Facility, Copenhagen. Observation year 2026. Accessed 2026-11-03.back
16. 5POST-CLOSE, context only: arXiv (ed.), Monthly submissions statistics (monthly CSV, 1991 to November 2026; annual totals summed by the editors). Cornell University, Ithaca. Observation year 2025. Accessed 2026-11-03.back
17. 26UNESCO (ed.), Recommendation on the Ethics of Artificial Intelligence, November 2021, programme page (193 Member States). UNESCO, Paris. Page read; the text itself was not opened. Observation year 2021. Accessed 2026-11-03.back
18. 28UNVERIFIED, not opened: UN General Assembly, resolution 79/325, Terms of reference and modalities for the establishment and functioning of the Independent International Scientific Panel on Artificial Intelligence and the Global Dialogue on Artificial Intelligence Governance, 2025; adoption in 2025 taken from search-result summaries, because the document page returned no text. United Nations, New York. Observation year 2025. Accessed 2026-11-03.back

5.1Assumptions

Scenario SCN-GS15-INT-01: what is assumed, and what is held constant.

Baseline

the 2026 reference state in this edition

Drivers

a fast-moving health or technology emergency in which many preprints appear, some produced at scale or with AI assistance, and some are retracted within weeks

Analogues

retraction waves of 2023⁶; paper-mill case studies¹⁴

Interventions

none modelled: the specification describes the unmitigated case

Run state

narrative specification; quantified run scheduled for 2027

5.2The cascade, step by step

1. 1

   DAT → GOV

   preprints and papers multiply faster than checks

   2023 to 2025

2. 2

   AIC → DAT

   AI-assisted text and mass production add low-quality items

   2025

3. 3

   DAT → HLT

   guidance cites results that are later retracted or not replicated

   2015 to 2023

4. 4

   GOV → FIN

   funders and markets react to reversals

   to be tested

5. 5

   DAT → TRD

   cross-border co-authors and platforms propagate corrections unevenly

   to be tested

6. 6

   GOV → DAT

   public trust in the record falls and reuse slows

   to be tested

5.3The scenario as a table of steps

How the shock in scenario SCN-GS15-INT-01 would travel through Global Science and Engineering Frontier’s coupled systems, step by step, under the assumptions the edition states.

A teaching model of a conditional scenario under the edition’s stated assumptions: not a forecast, not a prediction and not a warning. Sources: Global Science and Engineering Frontier 2026, scenario SCN-GS15-INT-01 (2026); Global Science and Engineering Frontier 2026, references (NR-CIT-1) (2026).

Notes

Numbers are the edition’s reference numbers, so a note reads the same in every chapter. The full list, with every source and its licence, is in Annex C. Annex C, references

1. 6Crossref (ed.), Crossref REST API, work counts by type and publication year and counts of retraction notices (update-type filter), queried live on the access date. Crossref, Oxford. Counts are subject to later registration. Observation year 2025 (series 2015 to 2025). Accessed 2026-11-03.back
2. 14Richardson, R.A.K. et al., The entities enabling scientific fraud at scale are large, resilient, and growing rapidly, Proceedings of the National Academy of Sciences 122(32), 2025 (abstract and open full text record read through Europe PMC). National Academy of Sciences, Washington. Observation year 2025. Accessed 2026-11-03.back

6.1The other desks’ editions of this subject

Federation: each desk writes its own edition for the subjects in its remit. Until a desk releases, its edition shows its frame and the reading this edition printed for it.

6.2Three lenses

One edition, read by each steward. The lens of this site comes first.

6.3Standards, Academy and steward contributions

How the Nexus Standards drafts, the Risk Academy and the three stewards feed each quarterly edition for science and engineering. The standards are a preparatory working draft (2026) and are cited as drafts, never as adopted standards.

6.4GCRI reading: the model card

GCRI’s reading of this edition is its model card: what the national model is, what it computes and where it stops. It is printed in full in Annex A, with the methods it belongs to. Annex A, methodology

6.5Global Risks Forum reading: editorial essays

Four signed essays by the Reports editorial team set out the positions this edition takes. They are interpretation, labelled as such; the evidence they draw on is in the findings register.

6.6Global Risks Alliance reading: 2027 science and engineering readiness pack

The registers that open the 2027 global cycle. At baseline each is initialised from this edition.

Publication dispositions

• Publish
• Hold
• Needs evidence
• Restricted
• Superseded

6.7Nexus Standards reading: standards and regulatory watch

Instruments and published guidance that shape the systems in this edition, by status on the retrieval date. Nexus standards drafts used for terms and evidence classes are cited as preparatory working draft (2026). The edition assesses no one’s compliance.

• ADOPTED
• PUBLISHED · GUIDANCE
• PUBLISHED · VOLUNTARY
• NOT VERIFIED

6.8Risk Academy reading: literature and evidence review, science and assessments

The review behind this edition covers science and engineering assessments, peer-reviewed science and the institutional reports that bind its findings. Systematic reviews for each agenda item begin in 2027.

Review protocol

Question

What does the published evidence establish about the world’s science and engineering enterprise and its couplings?

Sources searched

agency and academy reports; peer-reviewed science-of-science literature; statistics and open data services

Window

releases available at evidence close, 30 September 2026

Selection

sources that bind or contextualise a finding in this edition

Next step

systematic reviews for agenda items 01 to 05 in 2027

Notes

Numbers are the edition’s reference numbers, so a note reads the same in every chapter. The full list, with every source and its licence, is in Annex C. Annex C, references

1. 1NSB (ed.), Science and Engineering Indicators 2026: The State of U.S. Science and Engineering (NSB-SEP-2026-1), May 2026, National Science Board and National Center for Science and Engineering Statistics, US National Science Foundation, Alexandria. Full PDF opened; global R&D, publications, collaboration, venture investment, semiconductor and doctoral-degree figures read from the text. Observation year 2024. Accessed 2026-11-03.back
2. 24European Commission (ed.), Horizon Europe, programme page (indicative funding of EUR 93.5 billion for 2021 to 2027 after the mid-term review). European Commission, Brussels. Page read. Observation year 2026. Accessed 2026-11-03.back
3. 25UNESCO (ed.), UNESCO Recommendation on Open Science, 2021, programme page (193 Member States). UNESCO, Paris. Page read; the Recommendation text itself returned an access error. Observation year 2021. Accessed 2026-11-03.back
4. 26UNESCO (ed.), Recommendation on the Ethics of Artificial Intelligence, November 2021, programme page (193 Member States). UNESCO, Paris. Page read; the text itself was not opened. Observation year 2021. Accessed 2026-11-03.back
5. 37UNVERIFIED, not opened: UNESCO (ed.), Recommendation on Science and Scientific Researchers, 2017, cited from memory. UNESCO, Paris. Observation year 2017. Accessed 2026-11-03.back
6. 18WHO (ed.), Global guidance framework for the responsible use of the life sciences: mitigating biorisks and governing dual-use research, 174 pages, 2022. World Health Organization, Geneva. Publication page and overview read; the full text was not opened. Observation year 2022. Accessed 2026-11-03.back
7. 28UNVERIFIED, not opened: UN General Assembly, resolution 79/325, Terms of reference and modalities for the establishment and functioning of the Independent International Scientific Panel on Artificial Intelligence and the Global Dialogue on Artificial Intelligence Governance, 2025; adoption in 2025 taken from search-result summaries, because the document page returned no text. United Nations, New York. Observation year 2025. Accessed 2026-11-03.back
8. 29UNVERIFIED, not opened: UN General Assembly, resolution 79/1, Pact for the Future, including the Global Digital Compact, 2024, cited from memory. United Nations, New York. Observation year 2024. Accessed 2026-11-03.back
9. 30UNVERIFIED, not opened: Regulation (EU) 2024/1689 laying down harmonised rules on artificial intelligence, Official Journal of the European Union, 2024, cited from memory; the publication page returned a pending response. Observation year 2024. Accessed 2026-11-03.back
10. 31UNVERIFIED, not opened: Council of Europe, Framework Convention on Artificial Intelligence and Human Rights, Democracy and the Rule of Law, 2024, cited from memory. Council of Europe, Strasbourg. Observation year 2024. Accessed 2026-11-03.back
11. 32UNVERIFIED, not opened: ISO/IEC 42001:2023, Artificial intelligence management system; the catalogue page returned an access error. International Organization for Standardization, Geneva. Observation year 2023. Accessed 2026-11-03.back
12. 33UNVERIFIED, not opened: ISO/IEC 17025:2017, General requirements for the competence of testing and calibration laboratories, cited from memory. International Organization for Standardization, Geneva. Observation year 2017. Accessed 2026-11-03.back
13. 38UNVERIFIED, not opened: San Francisco Declaration on Research Assessment (DORA), 2012, cited from memory. DORA, San Francisco. Observation year 2012. Accessed 2026-11-03.back
14. 41UNVERIFIED, not opened: Convention on the Prohibition of the Development, Production and Stockpiling of Bacteriological (Biological) and Toxin Weapons and on their Destruction, 1972, cited from memory. United Nations, Geneva. Observation year 1972. Accessed 2026-11-03.back
15. 9Stanford HAI (ed.), The 2026 AI Index Report, Chapter 5 Science, chapter page with its highlights. Stanford Institute for Human-Centered AI, Stanford. Page read; the chapter PDF was not opened. Observation year 2025. Accessed 2026-11-03.back
16. 4TOP500 project (ed.), TOP500 list, 67th edition, June 2026 (release text read in full for the leading systems). TOP500, Prometeus GmbH. Observation year 2026. Accessed 2026-11-03.back
17. 14Richardson, R.A.K. et al., The entities enabling scientific fraud at scale are large, resilient, and growing rapidly, Proceedings of the National Academy of Sciences 122(32), 2025 (abstract and open full text record read through Europe PMC). National Academy of Sciences, Washington. Observation year 2025. Accessed 2026-11-03.back
18. 15Park, M., Leahey, E. and Funk, R.J., Papers and patents are becoming less disruptive over time, Nature 613, 138 to 144, 2023 (abstract read through Europe PMC; full text not opened). Springer Nature, London. Observation year 2023. Accessed 2026-11-03.back
19. 16Open Science Collaboration, Estimating the reproducibility of psychological science, Science 349(6251), aac4716, 2015 (abstract read through Europe PMC; full text not opened). American Association for the Advancement of Science, Washington. Observation year 2015. Accessed 2026-11-03.back
20. 11Abramson, J. et al., Accurate structure prediction of biomolecular interactions with AlphaFold 3, Nature 630, 493 to 500, 2024 (abstract and open full text read through Europe PMC). Springer Nature, London. Observation year 2024. Accessed 2026-11-03.back
21. 35UNVERIFIED, not opened: Jumper, J. et al., Highly accurate protein structure prediction with AlphaFold, Nature 596, 583 to 589, 2021, cited from memory. Springer Nature, London. No figure from it is used. Observation year 2021. Accessed 2026-11-03.back
22. 10Google Quantum AI and collaborators, Quantum error correction below the surface code threshold, arXiv 2408.13687 (full text opened) and Nature 638, 920 to 926, 2024. Observation year 2024. Accessed 2026-11-03.back
23. 13Sevilla, J. et al., Compute Trends Across Three Eras of Machine Learning, arXiv 2202.05924, 2022 (abstract page read). Cornell University, Ithaca. Observation year 2022. Accessed 2026-11-03.back
24. 34UNVERIFIED, not opened: Baker, M., 1,500 scientists lift the lid on reproducibility, Nature 533, 452 to 454, 2016; the article is behind a subscription and only its bibliographic record was checked. Springer Nature, London. No figure from it is used. Observation year 2016. Accessed 2026-11-03.back
25. 42UNVERIFIED, not opened: Bloom, N., Jones, C.I., Van Reenen, J. and Webb, M., Are Ideas Getting Harder to Find?, American Economic Review 110(4), 1104 to 1144, 2020, cited from memory. American Economic Association, Nashville. No figure from it is used. Observation year 2020. Accessed 2026-11-03.back
26. 43UNVERIFIED, not opened: Wuchty, S., Jones, B.F. and Uzzi, B., The Increasing Dominance of Teams in Production of Knowledge, Science 316(5827), 1036 to 1039, 2007, cited from memory. American Association for the Advancement of Science, Washington. No figure from it is used. Observation year 2007. Accessed 2026-11-03.back
27. 8Stanford HAI (ed.), The 2026 AI Index Report, overview page with the ten top takeaways. Stanford Institute for Human-Centered AI, Stanford. Page read; the report PDF was not opened. Observation year 2025. Accessed 2026-11-03.back
28. 36UNVERIFIED, not opened: UNESCO (ed.), UNESCO Science Report: Towards 2030, 2021 edition, cited from memory. UNESCO, Paris. No figure from it is used. Observation year 2021. Accessed 2026-11-03.back
29. 39UNVERIFIED, not opened: OECD (ed.), Frascati Manual 2015: Guidelines for Collecting and Reporting Data on Research and Experimental Development, cited from memory. OECD Publishing, Paris. Observation year 2015. Accessed 2026-11-03.back
30. 40UNVERIFIED, not opened: WIPO (ed.), World Intellectual Property Indicators 2025, cited from memory; the opened World Bank series draws on WIPO data. World Intellectual Property Organization, Geneva. No figure from it is used. Observation year 2024. Accessed 2026-11-03.back
31. 27Gottweis, J. et al., Accelerating scientific discovery with Co-Scientist, arXiv 2502.18864 (version 2, 2026; abstract read through the arXiv API). Cornell University, Ithaca. Observation year 2025. Accessed 2026-11-03.back
32. 2UNESCO Institute for Statistics (ed.), UIS data API, data version 20260507 (February 2026 data release): indicators EXPGDP.TOT, RESDEN.INHAB.TFTE and FRESP.THC, world and regional aggregates. UNESCO, Montreal. Data pulled and read. Observation year 2023. Accessed 2026-11-03.back
33. 3DATA VINTAGE AFTER CLOSE: World Bank (ed.), World Development Indicators API (series updated 2026-10-08, after the 30 September evidence close; cited as the position at access): GB.XPD.RSDV.GD.ZS, SP.POP.SCIE.RD.P6, IP.JRN.ARTC.SC, IP.PAT.RESD, IP.PAT.NRES, world values. World Bank, Washington. Data pulled and read. Observation year 2023 (patents 2024, researchers 2020). Accessed 2026-11-03.back
34. 6Crossref (ed.), Crossref REST API, work counts by type and publication year and counts of retraction notices (update-type filter), queried live on the access date. Crossref, Oxford. Counts are subject to later registration. Observation year 2025 (series 2015 to 2025). Accessed 2026-11-03.back
35. 5POST-CLOSE, context only: arXiv (ed.), Monthly submissions statistics (monthly CSV, 1991 to November 2026; annual totals summed by the editors). Cornell University, Ithaca. Observation year 2025. Accessed 2026-11-03.back
36. 7bioRxiv (ed.), bioRxiv and medRxiv content detail API, new-paper counts for 2024 and 2025. Cold Spring Harbor Laboratory, Cold Spring Harbor. Observation year 2025. Accessed 2026-11-03.back
37. 21Our World in Data (ed.), Yearly number of objects launched into outer space (data compiled from the UN Office for Outer Space Affairs, CSV read). Global Change Data Lab, Oxford. Observation year 2025. Accessed 2026-11-03.back
38. 44UNVERIFIED, not opened: UNESCO Institute for Statistics (ed.), R&D (SDG 9.5) data release, February 2025; the page returned a script challenge, and 1.95% of GDP in 2022 and 31.1% women among researchers were seen only in search-result summaries and are not used. UNESCO, Montreal. Observation year 2022. Accessed 2026-11-03.back
39. 19CERN (ed.), Future Circular Collider, feasibility study page (results published in spring 2025: 91 km ring, estimated cost CHF 15 billion). CERN, Geneva. Page read. Observation year 2025. Accessed 2026-11-03.back
40. 20Lawrence Livermore National Laboratory (ed.), Achieving Fusion Ignition, National Ignition Facility page (ignition reached on 5 December 2022). LLNL, Livermore. Page read. Observation year 2022. Accessed 2026-11-03.back
41. 22RCSB Protein Data Bank (ed.), Current holdings, list of entry identifiers (262,922 entries counted by the editors on the access date). Research Collaboratory for Structural Bioinformatics. Observation year 2026. Accessed 2026-11-03.back
42. 23GBIF (ed.), GBIF API, occurrence count and counts by year (3,890,294,224 records on the access date; 305,893,347 dated 2025). Global Biodiversity Information Facility, Copenhagen. Observation year 2026. Accessed 2026-11-03.back
43. 12Varadi, M. et al., AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences, Nucleic Acids Research 52 (database issue), 368 to 375, 2024 (abstract read through Europe PMC). Oxford University Press, Oxford. Observation year 2024. Accessed 2026-11-03.back

7.1Questions for the operational series

1. F-01

   How would a loss of trust in the research record during a fast-moving emergency travel through health advice, markets and public debate?

   INI-GS15-INT-01 · research integrity signals; first output: concurrency analysis for four information channels.

2. F-02

   Where do research capacity and compute concentrate, and which regions fall below any usable threshold?

   INI-GS15-CAP-02 · research capacity and compute atlas; first output: capacity table by regional group.

3. F-03

   How much of the literature is open, replicated and reusable, by field?

   INI-GS15-OPN-03 · open data and replication; first output: field-level open and replication table.

4. F-04

   Which large facilities have long lead times, single funders or single sites?

   INI-GS15-FAC-04 · large-facility register; first output: register of facilities with cost and schedule.

5. F-05

   Where does dual-use life science research have governance in practice, not only in guidance?

   INI-GS15-BIO-05 · dual-use governance map; first output: instrument and practice overlay.

6. F-06

   Which retraction and paper-mill signals can be read on a common basis across publishers?

   To be admitted in 2027; first output: data-sharing scoping note.

7. F-07

   How do researcher mobility and visa rules change who can work at the frontier?

   To be admitted in 2027; first output: scoping note.

8. F-08

   Which science capabilities have been tested in real shortfalls, and which are only planned?

   Science capability register; first output: rung evidence for 15 functions.

7.2Portfolio proposed for the Q1 2027 alignment

Five initiatives and one Universe mission candidate are proposed for the 2027 science and engineering portfolio. They are programme objects: no partner, funder or member is attached to them in this edition.

7.3The pathway

The science and engineering edition reads national science chapters, regional research-capacity evidence and the global edition through one governed evidence graph. At baseline the federation is designed; the Canada and global science chapters are drafted. This page states what exists today and what comes next.

1. 01

   National science chapters

   research system of each national edition

   DRAFTED · further national baselines

2. 02

   Regional capacity evidence

   shared facilities, networks and funders

   PLANNED · regional Baseline Editions

3. 03

   Science and engineering sector model

   the global research system of systems

   v0 SPECIFIED · v1 with federation evidence

4. 04

   Global and strategic links

   GLB · SR-13 Strategic Technology Infrastructure and Sovereign Readiness · XT-02 Quantum Technologies · XT-07 HPC and Advanced Computing

   PLANNED · strategic and technology baselines

5. 05

   Universe

   annual proving environment

   CANDIDATE MISSION · mandate in 2027

6. 06

   Global Science and Engineering Frontier Report

   this series

   BASELINE DRAFT · 2027 annual edition

7. 07

   Evidence and correction

   one graph, many projections

   OPEN · first correction cycle in 2027

7.4Universe 2027: evidence pack for the science and engineering candidate mission

1. 01

   Mandate

2. 02

   Core Build · 1 month

3. 03

   Action · 1 week

4. 04

   Evidence seal

5. 05

   Portfolio correction

6. 06

   Next mandate

Mission

ID

MSN-GS15-2027-01 · Trust shock in the research record Universe mission

Role spine

Leadership sponsor · Institutional anchor · Mission lead: drawn from national pathways

Receiving actors

to be identified with the mandate

7.5Handoffs and routing

Where the baseline’s findings go next. A handoff routes a finding for awareness, review or work; it transfers no authority.

7.6Feeds and signal register

Escalation routes

• none · keep monitoring
• research question
• project
• risk object
• annual publication
• signal report

Notes

Numbers are the edition’s reference numbers, so a note reads the same in every chapter. The full list, with every source and its licence, is in Annex C. Annex C, references

1. 6Crossref (ed.), Crossref REST API, work counts by type and publication year and counts of retraction notices (update-type filter), queried live on the access date. Crossref, Oxford. Counts are subject to later registration. Observation year 2025 (series 2015 to 2025). Accessed 2026-11-03.back
2. 4TOP500 project (ed.), TOP500 list, 67th edition, June 2026 (release text read in full for the leading systems). TOP500, Prometeus GmbH. Observation year 2026. Accessed 2026-11-03.back
3. 9Stanford HAI (ed.), The 2026 AI Index Report, Chapter 5 Science, chapter page with its highlights. Stanford Institute for Human-Centered AI, Stanford. Page read; the chapter PDF was not opened. Observation year 2025. Accessed 2026-11-03.back

8.1Science and health

Health depends on discovery: structures, trials, treatments and the rules that govern risky research. The 2023 genome-editing authorisation and the open structure archives show the link; the WHO framework shows the governance side.

Related editions

• GLB Global Systems
• GS-04 Health and One Health
• GS-12 Earth Observation and Planetary Intelligence
• GS-13 Cyber and Digital Systems
• GS-14 Policy, Institutions and Governance
• XT-02 Quantum Technologies

8.2Science and nature

Nature is measured through shared archives of records and structures. Their size is a strength, and their coverage and bias are the open questions.

Related editions

• GLB Global Systems
• GS-04 Health and One Health
• GS-12 Earth Observation and Planetary Intelligence
• GS-13 Cyber and Digital Systems
• GS-14 Policy, Institutions and Governance
• XT-02 Quantum Technologies

8.3Science, compute and the digital layer

The frontier is increasingly computational. Exascale machines, quantum memories and AI systems change who can do what, and how fast the record grows.

Related editions

• GLB Global Systems
• GS-04 Health and One Health
• GS-12 Earth Observation and Planetary Intelligence
• GS-13 Cyber and Digital Systems
• GS-14 Policy, Institutions and Governance
• XT-02 Quantum Technologies

8.4Science, finance and talent

Money and people decide where the frontier is. Spending and researcher density are both concentrated, and mobility is changing who works where.

Related editions

• GLB Global Systems
• GS-04 Health and One Health
• GS-12 Earth Observation and Planetary Intelligence
• GS-13 Cyber and Digital Systems
• GS-14 Policy, Institutions and Governance
• XT-02 Quantum Technologies

8.5Integrity, openness and governance

The record is shared infrastructure. Retractions, paper mills and low replication strain it, and open science and AI ethics recommendations set the rules around it.

Related editions

• GLB Global Systems
• GS-04 Health and One Health
• GS-12 Earth Observation and Planetary Intelligence
• GS-13 Cyber and Digital Systems
• GS-14 Policy, Institutions and Governance
• XT-02 Quantum Technologies

Notes

Numbers are the edition’s reference numbers, so a note reads the same in every chapter. The full list, with every source and its licence, is in Annex C. Annex C, references

1. 12Varadi, M. et al., AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences, Nucleic Acids Research 52 (database issue), 368 to 375, 2024 (abstract read through Europe PMC). Oxford University Press, Oxford. Observation year 2024. Accessed 2026-11-03.back
2. 17US Food and Drug Administration (ed.), FDA Approves First Gene Therapies to Treat Patients with Sickle Cell Disease, press announcement of 8 December 2023. FDA, Silver Spring. Page read. Observation year 2023. Accessed 2026-11-03.back
3. 18WHO (ed.), Global guidance framework for the responsible use of the life sciences: mitigating biorisks and governing dual-use research, 174 pages, 2022. World Health Organization, Geneva. Publication page and overview read; the full text was not opened. Observation year 2022. Accessed 2026-11-03.back
4. 11Abramson, J. et al., Accurate structure prediction of biomolecular interactions with AlphaFold 3, Nature 630, 493 to 500, 2024 (abstract and open full text read through Europe PMC). Springer Nature, London. Observation year 2024. Accessed 2026-11-03.back
5. 23GBIF (ed.), GBIF API, occurrence count and counts by year (3,890,294,224 records on the access date; 305,893,347 dated 2025). Global Biodiversity Information Facility, Copenhagen. Observation year 2026. Accessed 2026-11-03.back
6. 22RCSB Protein Data Bank (ed.), Current holdings, list of entry identifiers (262,922 entries counted by the editors on the access date). Research Collaboratory for Structural Bioinformatics. Observation year 2026. Accessed 2026-11-03.back
7. 4TOP500 project (ed.), TOP500 list, 67th edition, June 2026 (release text read in full for the leading systems). TOP500, Prometeus GmbH. Observation year 2026. Accessed 2026-11-03.back
8. 13Sevilla, J. et al., Compute Trends Across Three Eras of Machine Learning, arXiv 2202.05924, 2022 (abstract page read). Cornell University, Ithaca. Observation year 2022. Accessed 2026-11-03.back
9. 10Google Quantum AI and collaborators, Quantum error correction below the surface code threshold, arXiv 2408.13687 (full text opened) and Nature 638, 920 to 926, 2024. Observation year 2024. Accessed 2026-11-03.back
10. 9Stanford HAI (ed.), The 2026 AI Index Report, Chapter 5 Science, chapter page with its highlights. Stanford Institute for Human-Centered AI, Stanford. Page read; the chapter PDF was not opened. Observation year 2025. Accessed 2026-11-03.back
11. 1NSB (ed.), Science and Engineering Indicators 2026: The State of U.S. Science and Engineering (NSB-SEP-2026-1), May 2026, National Science Board and National Center for Science and Engineering Statistics, US National Science Foundation, Alexandria. Full PDF opened; global R&D, publications, collaboration, venture investment, semiconductor and doctoral-degree figures read from the text. Observation year 2024. Accessed 2026-11-03.back
12. 8Stanford HAI (ed.), The 2026 AI Index Report, overview page with the ten top takeaways. Stanford Institute for Human-Centered AI, Stanford. Page read; the report PDF was not opened. Observation year 2025. Accessed 2026-11-03.back
13. 24European Commission (ed.), Horizon Europe, programme page (indicative funding of EUR 93.5 billion for 2021 to 2027 after the mid-term review). European Commission, Brussels. Page read. Observation year 2026. Accessed 2026-11-03.back
14. 6Crossref (ed.), Crossref REST API, work counts by type and publication year and counts of retraction notices (update-type filter), queried live on the access date. Crossref, Oxford. Counts are subject to later registration. Observation year 2025 (series 2015 to 2025). Accessed 2026-11-03.back
15. 16Open Science Collaboration, Estimating the reproducibility of psychological science, Science 349(6251), aac4716, 2015 (abstract read through Europe PMC; full text not opened). American Association for the Advancement of Science, Washington. Observation year 2015. Accessed 2026-11-03.back
16. 25UNESCO (ed.), UNESCO Recommendation on Open Science, 2021, programme page (193 Member States). UNESCO, Paris. Page read; the Recommendation text itself returned an access error. Observation year 2021. Accessed 2026-11-03.back
17. 14Richardson, R.A.K. et al., The entities enabling scientific fraud at scale are large, resilient, and growing rapidly, Proceedings of the National Academy of Sciences 122(32), 2025 (abstract and open full text record read through Europe PMC). National Academy of Sciences, Washington. Observation year 2025. Accessed 2026-11-03.back
18. 26UNESCO (ed.), Recommendation on the Ethics of Artificial Intelligence, November 2021, programme page (193 Member States). UNESCO, Paris. Page read; the text itself was not opened. Observation year 2021. Accessed 2026-11-03.back

A.1Methods

The global baseline uses published values and four simple derivations. Each is shown in full so that any reader can reproduce it.

A.2Model card, Science and Engineering Sector Model v0

Model ID

MDL-GS15-SCIENCE · v0.1

Purpose

represent the global science and engineering system and its couplings to health, food, nature and the digital layer

Family

graph-based system of systems

Inputs

the cited sources of this edition; no proprietary data

Method

expert-specified coupling strengths (STRONG · MODERATE · WEAK) from assessments and analogue events

Validation

none yet, first validation against federation evidence in 2027

Uncertainty

qualitative; strengths are hypotheses, not measurements

Domain of validity

global scale; not for asset-, site- or country-level use

Reproducibility

strength table and sources published in this edition

A.3The baseline evidence cut

Every Nexus edition is published from a frozen evidence cut. The global baseline cut is documentary: it pins the exact edition of every source cited. Federation cuts begin with the first national Foundry cuts in 2027.

Cut ID

DC-SECTOR-SCIENCE-GLOBAL-2026-B0

Type

baseline documentary cut

Resolution policy

pinned · as of 30 September 2026

Sources pinned

44 cited sources, each at the edition and access date in the references

Model runs

none at baseline, analytical values are stated with their method

Digest

computed at freeze and recorded in the publication record

Supersedes

nothing, first cut of the series

Superseded by

the 2027 science and engineering annual cut

Lineage

1. 01

   source edition

2. 02

   pinned citation

3. 03

   finding

4. 04

   page

5. 05

   publication record

A.4Evidence age and coverage

A 2026 global edition rests on evidence observed in many different years. This page shows each time base, so that no single year label hides older evidence.

A.5Evidence cards, two anchor sources

Evidence cards to follow in 2027

• Crossref⁶: retraction notices by publisher and reason
• UNESCO²: national series behind the regional values
• AI Index⁹: data tables behind the science chapter
• TOP500⁴: November 2026 list and country totals

A.6Publication acceptance record

Editorial self-check at evidence close. Every applicable item carries a disposition; the remaining items are review, approval and deposit.

A.7Status words

A.8How the edition is made, released and corrected

Every edition is built from a sealed evidence cut, reviewed under three keys, signed and deposited before any site shows it as published; corrections go to Nexus Standards as change proposals and feed the next cut.

A teaching model of the process, not a forecast and not a record of any run; dates and counts on this page come from the programme record. Sources: Reports governance at GCRI Canada (2026); Zenodo, repository of record for Reports deposits (2026).

Lifecycle, one state at a time: Draft, Computed, Evidence bound, Reviewed, Approved for release, Published, Corrected, Superseded. This edition is a draft.

A.9Glossary and abbreviations

A.10Technical annex

A.1 · indicator definition: researchers per million inhabitants

Construct

full-time equivalent researchers engaged in R&D, per million inhabitants

Unit

researchers per million people

Method

issuer indicator based on national R&D surveys; regional aggregates

Source

UNESCO Institute for Statistics, February 2026 data release²

Uncertainty

aggregates depend on national coverage and the World Bank series differs

Known limits

measures headcount in effort, not quality, equipment or freedom to publish

Notes

Numbers are the edition’s reference numbers, so a note reads the same in every chapter. The full list, with every source and its licence, is in Annex C. Annex C, references

1. 2UNESCO Institute for Statistics (ed.), UIS data API, data version 20260507 (February 2026 data release): indicators EXPGDP.TOT, RESDEN.INHAB.TFTE and FRESP.THC, world and regional aggregates. UNESCO, Montreal. Data pulled and read. Observation year 2023. Accessed 2026-11-03.back
2. 3DATA VINTAGE AFTER CLOSE: World Bank (ed.), World Development Indicators API (series updated 2026-10-08, after the 30 September evidence close; cited as the position at access): GB.XPD.RSDV.GD.ZS, SP.POP.SCIE.RD.P6, IP.JRN.ARTC.SC, IP.PAT.RESD, IP.PAT.NRES, world values. World Bank, Washington. Data pulled and read. Observation year 2023 (patents 2024, researchers 2020). Accessed 2026-11-03.back
3. 39UNVERIFIED, not opened: OECD (ed.), Frascati Manual 2015: Guidelines for Collecting and Reporting Data on Research and Experimental Development, cited from memory. OECD Publishing, Paris. Observation year 2015. Accessed 2026-11-03.back
4. 14Richardson, R.A.K. et al., The entities enabling scientific fraud at scale are large, resilient, and growing rapidly, Proceedings of the National Academy of Sciences 122(32), 2025 (abstract and open full text record read through Europe PMC). National Academy of Sciences, Washington. Observation year 2025. Accessed 2026-11-03.back
5. 16Open Science Collaboration, Estimating the reproducibility of psychological science, Science 349(6251), aac4716, 2015 (abstract read through Europe PMC; full text not opened). American Association for the Advancement of Science, Washington. Observation year 2015. Accessed 2026-11-03.back
6. 34UNVERIFIED, not opened: Baker, M., 1,500 scientists lift the lid on reproducibility, Nature 533, 452 to 454, 2016; the article is behind a subscription and only its bibliographic record was checked. Springer Nature, London. No figure from it is used. Observation year 2016. Accessed 2026-11-03.back
7. 9Stanford HAI (ed.), The 2026 AI Index Report, Chapter 5 Science, chapter page with its highlights. Stanford Institute for Human-Centered AI, Stanford. Page read; the chapter PDF was not opened. Observation year 2025. Accessed 2026-11-03.back
8. 27Gottweis, J. et al., Accelerating scientific discovery with Co-Scientist, arXiv 2502.18864 (version 2, 2026; abstract read through the arXiv API). Cornell University, Ithaca. Observation year 2025. Accessed 2026-11-03.back
9. 4TOP500 project (ed.), TOP500 list, 67th edition, June 2026 (release text read in full for the leading systems). TOP500, Prometeus GmbH. Observation year 2026. Accessed 2026-11-03.back
10. 1NSB (ed.), Science and Engineering Indicators 2026: The State of U.S. Science and Engineering (NSB-SEP-2026-1), May 2026, National Science Board and National Center for Science and Engineering Statistics, US National Science Foundation, Alexandria. Full PDF opened; global R&D, publications, collaboration, venture investment, semiconductor and doctoral-degree figures read from the text. Observation year 2024. Accessed 2026-11-03.back
11. 6Crossref (ed.), Crossref REST API, work counts by type and publication year and counts of retraction notices (update-type filter), queried live on the access date. Crossref, Oxford. Counts are subject to later registration. Observation year 2025 (series 2015 to 2025). Accessed 2026-11-03.back

B.1The indicator set

B.2Global source register

The governed source set behind this edition. Each source keeps its authority, coverage, cadence and role; categories describe origin, not truth.

Domains bound by the register

• FIN
• DAT
• TRD
• GOV
• AIC
• INF
• BIO
• HLT
• MIN
• CLI

B.3Regional reading: science and engineering across ten clusters

Map plates follow with the regional Baseline Editions, when boundary sources and public-safe resolutions are bound. This page gives the regional reading in table form and ranks no region.

B.4Facility card: Future Circular Collider study

• facility
• accelerator
• long-lead
• international

Systems in view

research infrastructure, civil engineering, funding, data

Defining events

feasibility study published in spring 2025¹⁹

Couplings

INF → DAT data; INF → FIN cost; FIN → INF schedule

Capability evidence

a four-year study with national authorities and a preferred layout chosen from about 100 variants¹⁹

Gaps

funding commitments; competing proposals

Agenda items

F-04

B.5How countries appear in the science and engineering edition

Countries enter the science and engineering edition through the research chapters of their own national editions, on one card grammar. They are read against peer distributions and thresholds, never ranked.

Editions in the 2026 Baseline programme

How to read a country card

• Each system is shown against the distribution of its peers, with thresholds where they exist
• Couplings and risk objects come from the country’s own national edition
• Capability rungs record evidence, not a judgement of institutions
• Evidence age is printed, so older national data are visible as such

B.6Licences

B.7Formats and provenance

Provenance

Edition pack

The edition’s content pack (JSON), from which the print and web editions are built

Evidence cut

DC-SECTOR-SCIENCE-GLOBAL-2026-B0, closed 30 September 2026

Reproducibility

sourced values cited; analytical values reproducible from stated parameters

Self-check

14 of 23 at evidence close; a fail on any item blocks release

Review

five passes scheduled before release (in Annex C)

Publisher of record

GCRI for the Nexus Consortium

Corrections

Registry correction pathway; upheld corrections are published as notices with a new version

Notes

Numbers are the edition’s reference numbers, so a note reads the same in every chapter. The full list, with every source and its licence, is in Annex C. Annex C, references

1. 1NSB (ed.), Science and Engineering Indicators 2026: The State of U.S. Science and Engineering (NSB-SEP-2026-1), May 2026, National Science Board and National Center for Science and Engineering Statistics, US National Science Foundation, Alexandria. Full PDF opened; global R&D, publications, collaboration, venture investment, semiconductor and doctoral-degree figures read from the text. Observation year 2024. Accessed 2026-11-03.back
2. 2UNESCO Institute for Statistics (ed.), UIS data API, data version 20260507 (February 2026 data release): indicators EXPGDP.TOT, RESDEN.INHAB.TFTE and FRESP.THC, world and regional aggregates. UNESCO, Montreal. Data pulled and read. Observation year 2023. Accessed 2026-11-03.back
3. 3DATA VINTAGE AFTER CLOSE: World Bank (ed.), World Development Indicators API (series updated 2026-10-08, after the 30 September evidence close; cited as the position at access): GB.XPD.RSDV.GD.ZS, SP.POP.SCIE.RD.P6, IP.JRN.ARTC.SC, IP.PAT.RESD, IP.PAT.NRES, world values. World Bank, Washington. Data pulled and read. Observation year 2023 (patents 2024, researchers 2020). Accessed 2026-11-03.back
4. 5POST-CLOSE, context only: arXiv (ed.), Monthly submissions statistics (monthly CSV, 1991 to November 2026; annual totals summed by the editors). Cornell University, Ithaca. Observation year 2025. Accessed 2026-11-03.back
5. 6Crossref (ed.), Crossref REST API, work counts by type and publication year and counts of retraction notices (update-type filter), queried live on the access date. Crossref, Oxford. Counts are subject to later registration. Observation year 2025 (series 2015 to 2025). Accessed 2026-11-03.back
6. 4TOP500 project (ed.), TOP500 list, 67th edition, June 2026 (release text read in full for the leading systems). TOP500, Prometeus GmbH. Observation year 2026. Accessed 2026-11-03.back
7. 21Our World in Data (ed.), Yearly number of objects launched into outer space (data compiled from the UN Office for Outer Space Affairs, CSV read). Global Change Data Lab, Oxford. Observation year 2025. Accessed 2026-11-03.back
8. 22RCSB Protein Data Bank (ed.), Current holdings, list of entry identifiers (262,922 entries counted by the editors on the access date). Research Collaboratory for Structural Bioinformatics. Observation year 2026. Accessed 2026-11-03.back
9. 8Stanford HAI (ed.), The 2026 AI Index Report, overview page with the ten top takeaways. Stanford Institute for Human-Centered AI, Stanford. Page read; the report PDF was not opened. Observation year 2025. Accessed 2026-11-03.back
10. 23GBIF (ed.), GBIF API, occurrence count and counts by year (3,890,294,224 records on the access date; 305,893,347 dated 2025). Global Biodiversity Information Facility, Copenhagen. Observation year 2026. Accessed 2026-11-03.back
11. 18WHO (ed.), Global guidance framework for the responsible use of the life sciences: mitigating biorisks and governing dual-use research, 174 pages, 2022. World Health Organization, Geneva. Publication page and overview read; the full text was not opened. Observation year 2022. Accessed 2026-11-03.back
12. 24European Commission (ed.), Horizon Europe, programme page (indicative funding of EUR 93.5 billion for 2021 to 2027 after the mid-term review). European Commission, Brussels. Page read. Observation year 2026. Accessed 2026-11-03.back
13. 19CERN (ed.), Future Circular Collider, feasibility study page (results published in spring 2025: 91 km ring, estimated cost CHF 15 billion). CERN, Geneva. Page read. Observation year 2025. Accessed 2026-11-03.back

C.144 sources

NR-CIT-1, editor first. A DOI is linked where the source has one; Nexus records carry a UNI once the catalogue allocates it (NS-1).

1. 1NSB (ed.), Science and Engineering Indicators 2026: The State of U.S. Science and Engineering (NSB-SEP-2026-1), May 2026, National Science Board and National Center for Science and Engineering Statistics, US National Science Foundation, Alexandria. Full PDF opened; global R&D, publications, collaboration, venture investment, semiconductor and doctoral-degree figures read from the text. Observation year 2024. Accessed 2026-11-03. Cited in the summary, chapter 1, chapter 2, chapter 3, chapter 4, chapter 6, chapter 8
2. 2UNESCO Institute for Statistics (ed.), UIS data API, data version 20260507 (February 2026 data release): indicators EXPGDP.TOT, RESDEN.INHAB.TFTE and FRESP.THC, world and regional aggregates. UNESCO, Montreal. Data pulled and read. Observation year 2023. Accessed 2026-11-03. Cited in the summary, chapter 1, chapter 2, chapter 3, chapter 4, chapter 6
3. 3DATA VINTAGE AFTER CLOSE: World Bank (ed.), World Development Indicators API (series updated 2026-10-08, after the 30 September evidence close; cited as the position at access): GB.XPD.RSDV.GD.ZS, SP.POP.SCIE.RD.P6, IP.JRN.ARTC.SC, IP.PAT.RESD, IP.PAT.NRES, world values. World Bank, Washington. Data pulled and read. Observation year 2023 (patents 2024, researchers 2020). Accessed 2026-11-03. Cited in the summary, chapter 2, chapter 6
4. 4TOP500 project (ed.), TOP500 list, 67th edition, June 2026 (release text read in full for the leading systems). TOP500, Prometeus GmbH. Observation year 2026. Accessed 2026-11-03. Cited in the summary, chapter 1, chapter 2, chapter 3, chapter 4, chapter 6, chapter 7, chapter 8
5. 5POST-CLOSE, context only: arXiv (ed.), Monthly submissions statistics (monthly CSV, 1991 to November 2026; annual totals summed by the editors). Cornell University, Ithaca. Observation year 2025. Accessed 2026-11-03. Cited in the summary, chapter 1, chapter 4, chapter 6
6. 6Crossref (ed.), Crossref REST API, work counts by type and publication year and counts of retraction notices (update-type filter), queried live on the access date. Crossref, Oxford. Counts are subject to later registration. Observation year 2025 (series 2015 to 2025). Accessed 2026-11-03. Cited in the summary, chapter 1, chapter 2, chapter 3, chapter 4, chapter 5, chapter 6, chapter 7, chapter 8
7. 7bioRxiv (ed.), bioRxiv and medRxiv content detail API, new-paper counts for 2024 and 2025. Cold Spring Harbor Laboratory, Cold Spring Harbor. Observation year 2025. Accessed 2026-11-03. Cited in chapter 6
8. 8Stanford HAI (ed.), The 2026 AI Index Report, overview page with the ten top takeaways. Stanford Institute for Human-Centered AI, Stanford. Page read; the report PDF was not opened. Observation year 2025. Accessed 2026-11-03. Cited in the summary, chapter 1, chapter 2, chapter 6, chapter 8
9. 9Stanford HAI (ed.), The 2026 AI Index Report, Chapter 5 Science, chapter page with its highlights. Stanford Institute for Human-Centered AI, Stanford. Page read; the chapter PDF was not opened. Observation year 2025. Accessed 2026-11-03. Cited in the summary, chapter 1, chapter 2, chapter 3, chapter 4, chapter 6, chapter 7, chapter 8
10. 10Google Quantum AI and collaborators, Quantum error correction below the surface code threshold, arXiv 2408.13687 (full text opened) and Nature 638, 920 to 926, 2024. Observation year 2024. Accessed 2026-11-03. Cited in the summary, chapter 2, chapter 6, chapter 8
11. 11Abramson, J. et al., Accurate structure prediction of biomolecular interactions with AlphaFold 3, Nature 630, 493 to 500, 2024 (abstract and open full text read through Europe PMC). Springer Nature, London. Observation year 2024. Accessed 2026-11-03. Cited in chapter 1, chapter 3, chapter 4, chapter 6, chapter 8
12. 12Varadi, M. et al., AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences, Nucleic Acids Research 52 (database issue), 368 to 375, 2024 (abstract read through Europe PMC). Oxford University Press, Oxford. Observation year 2024. Accessed 2026-11-03. Cited in the summary, chapter 1, chapter 2, chapter 6, chapter 8
13. 13Sevilla, J. et al., Compute Trends Across Three Eras of Machine Learning, arXiv 2202.05924, 2022 (abstract page read). Cornell University, Ithaca. Observation year 2022. Accessed 2026-11-03. Cited in chapter 1, chapter 2, chapter 6, chapter 8
14. 14Richardson, R.A.K. et al., The entities enabling scientific fraud at scale are large, resilient, and growing rapidly, Proceedings of the National Academy of Sciences 122(32), 2025 (abstract and open full text record read through Europe PMC). National Academy of Sciences, Washington. Observation year 2025. Accessed 2026-11-03. Cited in the summary, chapter 1, chapter 2, chapter 4, chapter 5, chapter 6, chapter 8
15. 15Park, M., Leahey, E. and Funk, R.J., Papers and patents are becoming less disruptive over time, Nature 613, 138 to 144, 2023 (abstract read through Europe PMC; full text not opened). Springer Nature, London. Observation year 2023. Accessed 2026-11-03. Cited in chapter 6
16. 16Open Science Collaboration, Estimating the reproducibility of psychological science, Science 349(6251), aac4716, 2015 (abstract read through Europe PMC; full text not opened). American Association for the Advancement of Science, Washington. Observation year 2015. Accessed 2026-11-03. Cited in chapter 1, chapter 4, chapter 6, chapter 8
17. 17US Food and Drug Administration (ed.), FDA Approves First Gene Therapies to Treat Patients with Sickle Cell Disease, press announcement of 8 December 2023. FDA, Silver Spring. Page read. Observation year 2023. Accessed 2026-11-03. Cited in chapter 1, chapter 2, chapter 3, chapter 8
18. 18WHO (ed.), Global guidance framework for the responsible use of the life sciences: mitigating biorisks and governing dual-use research, 174 pages, 2022. World Health Organization, Geneva. Publication page and overview read; the full text was not opened. Observation year 2022. Accessed 2026-11-03. Cited in chapter 1, chapter 2, chapter 3, chapter 4, chapter 6, chapter 8
19. 19CERN (ed.), Future Circular Collider, feasibility study page (results published in spring 2025: 91 km ring, estimated cost CHF 15 billion). CERN, Geneva. Page read. Observation year 2025. Accessed 2026-11-03. Cited in chapter 3, chapter 4, chapter 6
20. 20Lawrence Livermore National Laboratory (ed.), Achieving Fusion Ignition, National Ignition Facility page (ignition reached on 5 December 2022). LLNL, Livermore. Page read. Observation year 2022. Accessed 2026-11-03. Cited in chapter 6
21. 21Our World in Data (ed.), Yearly number of objects launched into outer space (data compiled from the UN Office for Outer Space Affairs, CSV read). Global Change Data Lab, Oxford. Observation year 2025. Accessed 2026-11-03. Cited in chapter 6
22. 22RCSB Protein Data Bank (ed.), Current holdings, list of entry identifiers (262,922 entries counted by the editors on the access date). Research Collaboratory for Structural Bioinformatics. Observation year 2026. Accessed 2026-11-03. Cited in the summary, chapter 2, chapter 4, chapter 6, chapter 8
23. 23GBIF (ed.), GBIF API, occurrence count and counts by year (3,890,294,224 records on the access date; 305,893,347 dated 2025). Global Biodiversity Information Facility, Copenhagen. Observation year 2026. Accessed 2026-11-03. Cited in the summary, chapter 2, chapter 3, chapter 4, chapter 6, chapter 8
24. 24European Commission (ed.), Horizon Europe, programme page (indicative funding of EUR 93.5 billion for 2021 to 2027 after the mid-term review). European Commission, Brussels. Page read. Observation year 2026. Accessed 2026-11-03. Cited in chapter 4, chapter 6, chapter 8
25. 25UNESCO (ed.), UNESCO Recommendation on Open Science, 2021, programme page (193 Member States). UNESCO, Paris. Page read; the Recommendation text itself returned an access error. Observation year 2021. Accessed 2026-11-03. Cited in chapter 1, chapter 3, chapter 4, chapter 6, chapter 8
26. 26UNESCO (ed.), Recommendation on the Ethics of Artificial Intelligence, November 2021, programme page (193 Member States). UNESCO, Paris. Page read; the text itself was not opened. Observation year 2021. Accessed 2026-11-03. Cited in chapter 1, chapter 4, chapter 6, chapter 8
27. 27Gottweis, J. et al., Accelerating scientific discovery with Co-Scientist, arXiv 2502.18864 (version 2, 2026; abstract read through the arXiv API). Cornell University, Ithaca. Observation year 2025. Accessed 2026-11-03. Cited in chapter 1, chapter 6
28. 28UNVERIFIED, not opened: UN General Assembly, resolution 79/325, Terms of reference and modalities for the establishment and functioning of the Independent International Scientific Panel on Artificial Intelligence and the Global Dialogue on Artificial Intelligence Governance, 2025; adoption in 2025 taken from search-result summaries, because the document page returned no text. United Nations, New York. Observation year 2025. Accessed 2026-11-03. Cited in chapter 4, chapter 6
29. 29UNVERIFIED, not opened: UN General Assembly, resolution 79/1, Pact for the Future, including the Global Digital Compact, 2024, cited from memory. United Nations, New York. Observation year 2024. Accessed 2026-11-03. Cited in chapter 6
30. 30UNVERIFIED, not opened: Regulation (EU) 2024/1689 laying down harmonised rules on artificial intelligence, Official Journal of the European Union, 2024, cited from memory; the publication page returned a pending response. Observation year 2024. Accessed 2026-11-03. Cited in chapter 6
31. 31UNVERIFIED, not opened: Council of Europe, Framework Convention on Artificial Intelligence and Human Rights, Democracy and the Rule of Law, 2024, cited from memory. Council of Europe, Strasbourg. Observation year 2024. Accessed 2026-11-03. Cited in chapter 6
32. 32UNVERIFIED, not opened: ISO/IEC 42001:2023, Artificial intelligence management system; the catalogue page returned an access error. International Organization for Standardization, Geneva. Observation year 2023. Accessed 2026-11-03. Cited in chapter 6
33. 33UNVERIFIED, not opened: ISO/IEC 17025:2017, General requirements for the competence of testing and calibration laboratories, cited from memory. International Organization for Standardization, Geneva. Observation year 2017. Accessed 2026-11-03. Cited in chapter 6
34. 34UNVERIFIED, not opened: Baker, M., 1,500 scientists lift the lid on reproducibility, Nature 533, 452 to 454, 2016; the article is behind a subscription and only its bibliographic record was checked. Springer Nature, London. No figure from it is used. Observation year 2016. Accessed 2026-11-03. Cited in chapter 6
35. 35UNVERIFIED, not opened: Jumper, J. et al., Highly accurate protein structure prediction with AlphaFold, Nature 596, 583 to 589, 2021, cited from memory. Springer Nature, London. No figure from it is used. Observation year 2021. Accessed 2026-11-03. Cited in chapter 6
36. 36UNVERIFIED, not opened: UNESCO (ed.), UNESCO Science Report: Towards 2030, 2021 edition, cited from memory. UNESCO, Paris. No figure from it is used. Observation year 2021. Accessed 2026-11-03. Cited in chapter 6
37. 37UNVERIFIED, not opened: UNESCO (ed.), Recommendation on Science and Scientific Researchers, 2017, cited from memory. UNESCO, Paris. Observation year 2017. Accessed 2026-11-03. Cited in chapter 4, chapter 6
38. 38UNVERIFIED, not opened: San Francisco Declaration on Research Assessment (DORA), 2012, cited from memory. DORA, San Francisco. Observation year 2012. Accessed 2026-11-03. Cited in chapter 6
39. 39UNVERIFIED, not opened: OECD (ed.), Frascati Manual 2015: Guidelines for Collecting and Reporting Data on Research and Experimental Development, cited from memory. OECD Publishing, Paris. Observation year 2015. Accessed 2026-11-03. Cited in chapter 6
40. 40UNVERIFIED, not opened: WIPO (ed.), World Intellectual Property Indicators 2025, cited from memory; the opened World Bank series draws on WIPO data. World Intellectual Property Organization, Geneva. No figure from it is used. Observation year 2024. Accessed 2026-11-03. Cited in chapter 6
41. 41UNVERIFIED, not opened: Convention on the Prohibition of the Development, Production and Stockpiling of Bacteriological (Biological) and Toxin Weapons and on their Destruction, 1972, cited from memory. United Nations, Geneva. Observation year 1972. Accessed 2026-11-03. Cited in chapter 6
42. 42UNVERIFIED, not opened: Bloom, N., Jones, C.I., Van Reenen, J. and Webb, M., Are Ideas Getting Harder to Find?, American Economic Review 110(4), 1104 to 1144, 2020, cited from memory. American Economic Association, Nashville. No figure from it is used. Observation year 2020. Accessed 2026-11-03. Cited in chapter 6
43. 43UNVERIFIED, not opened: Wuchty, S., Jones, B.F. and Uzzi, B., The Increasing Dominance of Teams in Production of Knowledge, Science 316(5827), 1036 to 1039, 2007, cited from memory. American Association for the Advancement of Science, Washington. No figure from it is used. Observation year 2007. Accessed 2026-11-03. Cited in chapter 6
44. 44UNVERIFIED, not opened: UNESCO Institute for Statistics (ed.), R&D (SDG 9.5) data release, February 2025; the page returned a script challenge, and 1.95% of GDP in 2022 and 31.1% women among researchers were seen only in search-result summaries and are not used. UNESCO, Montreal. Observation year 2022. Accessed 2026-11-03. Cited in chapter 6

C.2Publication record

Colophon

Format

A4 portrait · screen, print and archival PDF

Typefaces

Space Grotesk (display) · Geist (text) · IBM Plex Sans Italic (standfirsts) · IBM Plex Mono (data)

Design standard

Reports Foundational Edition, Volume I, Part VII

Accessibility

tagged PDF and HTML targets; text alternatives for every figure

Production

Reports editorial team, GCRI

Release path for this version