Decade of Open Data Evolution Unveiled in Latest Report

The 10th edition of the State of Open Data Report, titled The State of Open Data 2025: A Decade of Progress and Challenges, marks a significant milestone in understanding how researchers worldwide, including those in European higher education institutions, have embraced open data practices over the past ten years. Published by Digital Science, Figshare, and Springer Nature, this annual survey—now the longest-running study of its kind—draws from over 4,700 responses in 2025 alone, aggregating insights from more than 43,000 researchers across 212 countries since 2016. For European academics, the findings underscore a continent leading in policy-driven open science, yet facing familiar hurdles in implementation and recognition.

European universities have been at the forefront of this shift, influenced by initiatives like the European Open Science Cloud (EOSC) and Horizon Europe mandates that require data management plans emphasizing findability, accessibility, interoperability, and reusability—collectively known as the FAIR principles (Findable, Accessible, Interoperable, Reusable). The report reveals how these policies have shaped behaviors, with steady progress amid global divergences.

Tracking Key Metrics: From 2016 to 2025

Over the decade, the report monitors four pivotal metrics: awareness of FAIR principles, support for national data-sharing mandates, use of AI tools in data workflows, and perceptions of credit for sharing data. In Europe, FAIR awareness has skyrocketed, with unfamiliarity rates plummeting by 40-60 percentage points since 2018 across disciplines. For instance, fields like business and investment saw unfamiliarity drop from 80% to just 20%, reflecting broader institutional training in universities from the University of Manchester to ETH Zurich.

Support for mandates remains relatively stable in key European nations—Germany shows no significant decline, while Italy dipped only 11 points—contrasting sharp drops in Australia (35.8 points) and the US (23.2 points). This resilience highlights Europe's robust policy ecosystem, where funders like the European Research Council enforce open data from grant inception. Meanwhile, AI adoption for data processing jumped from 22.1% in 2024 to 31.9% in 2025, and metadata creation rose from 16.1% to 25.1%, signaling how tools are streamlining compliance in research-heavy environments like French Grandes Écoles or Dutch technical universities.

FAIR Principles: Mainstreamed in European Research

The FAIR principles, first articulated in 2016, have transitioned from niche concept to cornerstone of research integrity. The report documents a tripling in familiarity from 15.2% in 2018 to 40.6% in 2025 globally, but Europe's universities report even higher rates due to mandatory training. At institutions like the University of Edinburgh or KU Leuven, data stewards now embed FAIR in PhD programs, ensuring datasets are indexed with persistent identifiers and rich metadata.

Step-by-step, FAIR implementation involves: (1) assigning DOIs for findability; (2) using standard protocols for access; (3) adopting vocabularies like Dublin Core for interoperability; and (4) providing licenses and documentation for reusability. European case studies, such as the ELIXIR infrastructure for life sciences, demonstrate 89% public data availability rates, far exceeding global averages. Yet, challenges persist in humanities and social sciences, where qualitative data resists standardization.

AI's Accelerating Role in Open Data Workflows

Artificial Intelligence (AI) is reshaping data handling, with European researchers leading experimentation. The 10-point surge in active AI use for processing reflects tools like automated metadata generators and discrepancy detectors. In Germany’s Max Planck Society, AI normalizes formats across repositories, cutting manual labor by 30-50%.

• Benefits: Faster quality checks, enhanced interoperability, reduced barriers for early-career researchers.
• Risks: Bias amplification if training data lacks diversity; ethical concerns in sensitive European health datasets under GDPR.
• Comparisons: Machine learning conferences now normatively release code/datasets pre-review, boosting reusability.

As Dawei Zhang notes, future AI could auto-correct discrepancies with minimal intervention, vital for Europe's interdisciplinary collaborations.

Photo by Leonhard Niederwimmer on Unsplash

The Enduring Credit Gap Hindering Progress

Despite advances, 69.2% of researchers feel under-credited for data sharing—a slight narrowing from prior years but a stubborn barrier. In European higher education, promotion criteria still prioritize journal impact factors over data citations. Joy Owango emphasizes aligning incentives: "University leadership needs to reward open science like top-tier publications."

This gap disproportionately affects early-career academics in Spain or Poland, where funding ties to outputs. Solutions include citable data badges and ORCID integration, piloted at Scandinavian universities.

Europe's Unique Position Amid Global Divergences

While global mandate support wanes, Europe's stability—exemplified by Germany's consistency and Italy's resilience—stems from nuanced policies balancing mandates with infrastructure. The EU's Open Data Directive and national portals like data.gov.uk foster reuse, with maturity scores rising 3 points to 86% for EU-27 in related public data assessments.

Stakeholder views vary: Librarians advocate skills investment; funders push standards. Implications include boosted AI-driven insights from high-quality European datasets, but risks of 'data silos' in non-English languages.

Impacts on European Higher Education Institutions

Universities like Sorbonne or Heidelberg now integrate open data into curricula, preparing students for FAIR-compliant careers. Statistics show 80.9% researcher backing for open data, aligning with 88.1% for open access. Case: France's CNRS repositories enable cross-border reanalysis, accelerating discoveries in climate modeling.

Challenges: Resource strains in Eastern Europe; solutions via EOSC hubs providing shared storage. For academics eyeing research jobs, open data skills are premium.

Expert Insights and Stakeholder Perspectives

Brian Nosek frames open data as long-term investment: reanalysis may yield value years later. Hilary Hanahoe calls for skills to leverage AI with open data. In Europe, RDA Europe's work coordinates standards, ensuring cultural contexts like multilingual metadata.

Timelines: 2016 FAIR birth; 2020 COVID accelerates sharing; 2025 AI boom. Future: AI-enhanced interoperability.

Photo by Lysander Yuen on Unsplash

Pathways Forward: Recommendations for Europe

The report urges: (1) Reform assessments for citable data; (2) AI for metadata/quality; (3) Regional coordination via EOSC. European universities can lead by piloting badge systems and training. Actionable: Adopt academic CV templates highlighting data DOIs.

• Steps for unis: Mandate DMPs, fund stewards.
• Benefits: Reproducible science, funding edges.

Conclusion: Europe's Open Data Leadership Opportunity

The 10th State of Open Data Report positions Europe as a beacon amid global stalls. By addressing credit and investing in AI, universities can unlock data's full potential. Researchers, explore Rate My Professor for open science mentors, browse higher ed jobs, and access career advice. Institutions, post openings at University Jobs or Post a Job. The next decade awaits transformative reuse.