Spain leads global research on AI and journalism
February 13, 2026 - Image credit: Clarot & AI4Media (Better Images of AI)
Spain now tops global output on AI-and-journalism research. A systematic review by the Universitat Autònoma de Barcelona (UAB) and CEU Abat Oliba analyzed 203 English-language papers from 2020-2024 and found Spanish authors produced 26% of them. The work appears in the journal Review of Communication Research (DOI: 10.52152/rcr.v14.3).
Key findings
- Spain leads output: 26% of all papers, nearly double the U.S., nearly triple China. Six of the ten most productive countries are in Europe.
- Fast growth: 13 papers in 2020 vs. 102 in 2024 (half of the five-year total).
- Regional skew: Over half of studies (106) focus on Europe. Far fewer examine the Middle East (20), Latin America (17), or sub-Saharan Africa (10).
- Quality venues: 39% of studies appear in top-quartile journals.
What the field overlooks
The review flags persistent blind spots that limit practical guidance for newsrooms and policymakers. Much of the literature treats AI as an inevitable upgrade rather than a choice with trade-offs.
- Environmental impact: Energy use, emissions, and lifecycle costs of AI in news workflows are rarely quantified.
- Global South: Sparse evidence on deployment realities, constraints, and outcomes outside Europe and North America.
- Editorial judgment: Limited study of how algorithms influence news selection, sourcing, and verification.
- Ethics in practice: Many papers discuss bias and codes, but few test whether ethical guidelines or self-regulation actually work.
- Audience perception: Minimal research on trust, disclosure, and acceptance of AI-assisted news production.
Why this matters
AI adoption in newsrooms is outpacing evidence on its real costs and benefits. Without coverage of these gaps, decision-makers default to hype, not data. The review argues for a shift toward work that tests outcomes, measures externalities, and centers audience trust.
Practical next steps for researchers and editors
- Measure environmental costs: Run energy and emissions audits for model training and inference; report methods and baselines.
- Study editorial effects: Evaluate how recommender systems and AI assistants affect gatekeeping, sourcing diversity, and error rates.
- Test ethics in the wild: Track compliance and effectiveness of newsroom AI policies; define clear success metrics and failure modes.
- Center audiences: Use experiments and panel data to assess labels, transparency notices, and their impact on trust and engagement.
- Rebalance geography: Fund comparative studies in the Global South with local co-authorship, field data, and multilingual designs.
- Open science: Share code, prompts, models, and datasets where possible; preregister studies to reduce bias.
Method notes
The review covers 203 English-language articles published from 2020 to 2024. Most were European in origin and scope. The study forms part of the IA-COM project on AI's role in quality journalism and media literacy amid disinformation.
Citation
More information: "Deconstructing the Hype: A Critical Literature Review on AI in Journalism," Review of Communication Research (2026). DOI: 10.52152/rcr.v14.3
Useful resources
If you're building AI literacy for research teams or newsroom training, see curated course pathways at Complete AI Training - Courses by Job or browse the latest AI courses.
Your membership also unlocks:
