How should journalism education change to respond to artificial intelligence? An international research programme led by the UNESCO Chair on Media and Information Literacy for Quality Journalism at the Universitat Autònoma de Barcelona will build scenarios for journalism up to 2036 and compare what they would demand of graduates with what undergraduate and master’s programmes teach today.

Artificial intelligence is changing both how journalists produce news and how people encounter it. Newsrooms can use AI to transcribe interviews, search archives or analyse large collections of documents. Meanwhile, AI assistants can offer answers about public affairs directly, sometimes leaving readers with little indication of who did the original reporting or how to check it.

The Future of Journalism Education in the Age of AI is an international research programme promoted by the UNESCO Chair on Media and Information Literacy for Quality Journalism at the Universitat Autònoma de Barcelona (UAB), in collaboration with the Gabinete de Comunicación y Educación of the same university, and within the framework of the UNESCO UNITWIN Network on Media and Information Literacy and Intercultural Dialogue (MILID). The programme aims to anticipate possible scenarios for journalism up to 2036 and use them to develop guidance for journalism education, with the broader goal of strengthening a reliable, plural information ecosystem grounded in freedom of expression and the right to information.

What the programme will investigate

The study connects three areas. The first is the transformation of the news sector, including changes in employment and professional roles, newsroom workflows, business models and the relationship between news organisations, technology providers and audiences. The second concerns expectations about AI. What do journalists, educators and students think is likely to happen, what would they prefer to happen, and how much capacity do they believe their institutions have to respond? The third concerns the competences that journalists and citizens need in an information environment increasingly mediated by AI.

Three tensions, already visible in today’s journalism, organise both the analysis and the scenarios. The first, between spectacularisation and understanding, concerns content: AI can multiply the staging of what governments, companies, parties and lobbies say, or free up resources to investigate what they do. The second, between concealment and transparency, concerns technological mediation: whether people can trace a claim to its sources and know who is responsible for checking or correcting it. The third, between fragmentation and integration, concerns the public sphere: whether reliable reporting reaches every community, or some are left in information environments dominated by entertainment and vulnerable to manipulation. None of these tensions has a predetermined outcome. All teams will track them with shared indicators, such as the share of stories based on statements rather than original reporting, or how accurately AI answers attribute their sources.

How the research will work

The study works at several levels. The basic unit is the country, because degrees are accredited, and labour markets and media systems largely organised, at national level; each national team forms a node. Countries are grouped into linguistic and cultural areas and wider regions, and are chosen to maximise variation in media systems, AI adoption, regulation, language and models of journalism education. Scenarios are always formulated for a defined geographical scope, and a common minimum core of instruments keeps the results comparable.

The provisional methodology then proposes a sequence of four linked tasks. A scoping review will bring together academic research, institutional reports and evidence from different regions and languages. Trend analysis and a Delphi consultation with an international expert panel will assess the likely direction of each tension. Treating each tension as an axis with two poles yields eight possible configurations, from which four contrasting and plausible scenarios up to 2036 will be developed, each stating what drives it and which early signals would show that it is taking shape. The scenarios will give researchers and educators a basis for testing which educational choices make sense across several different futures.

Alongside the scenario work, participating teams will map selected undergraduate and master’s programmes in journalism using a common curriculum record sheet. The analysis will examine where AI and media and information literacy appear in learning outcomes, teaching activities and assessment. It will distinguish practical tool use from critical understanding, professional ethics and strategic questions about how news organisations are changing. Questionnaires and interviews with educators will help establish how far formal course descriptions reflect what is taught and assessed.

The heart of the programme is the comparison. The competences that journalism will need are identified through three routes: demand from the sector (job profiles, vacancies and interviews with newsroom managers), anticipation (the scenarios and the Delphi consultation) and international reference frameworks, such as UNESCO’s AI competency frameworks for teachers and students (2024). The results are then set out in comparative tables showing, for each tension and each competence, what the scenarios anticipate, what curricula offer and what educators and students think. The gaps, and the competences that keep their value in any scenario, emerge from that cross-tabulation.

Educators and students will then discuss the scenarios and compare the futures they consider probable with those they would prefer. Workshops will use that comparison to identify changes that schools could begin to make now. National and regional teams will apply a common set of research instruments while documenting local differences in language, higher education, labour markets and media systems.

What earlier studies show

Two foresight reports illustrate the value of developing several plausible futures. David Caswell and Shuwei Fang’s AI in Journalism Futures (Open Society Foundations, 2024) used a participatory process to develop five scenarios for AI’s long-term influence on journalism and civic information. Media 2035, published by the Media Industry Research Foundation of Finland in 2024, explored four futures for the broader media sector, including a resurgence of demand for reliable journalism and a future in which entertainment further divides audiences. Open Society Foundations report · Media 2035 report

Two empirical reports help specify what should be measured. The Reuters Institute’s Generative AI and News Report 2026 found that weekly use of standalone generative AI tools and chatbots rose from 34% to 51% in a single year across the six countries for which it has earlier data. Across those same six countries, 9% reported using generative AI to get news in the previous week in 2026, up from 3% in 2024. A 2025 study by 22 public service media organisations in 18 countries, coordinated by the European Broadcasting Union and led by the BBC, assessed more than 3,000 responses from four AI assistants in 14 languages. Professional journalists found at least one significant issue in 45% of them. For journalism schools, these studies point to the value of teaching students how to audit an AI-generated answer, establish where its claims came from and decide what evidence a reader would need to verify them. Reuters Institute report · EBU and BBC report

Research on journalism education offers a starting point for the curriculum analysis. In 2024, Santiago Tejedor and colleagues reviewed the journalism degrees of 41 Spanish universities: of 1,775 courses, seven explicitly covered data journalism or big data, and 19 included AI-related topics within broader course content. Debora Wenger and colleagues, drawing on interviews with leaders of accredited journalism programmes in the United States, identified faculty expertise, institutional policies and ethical questions as practical challenges. Together these studies show why it is worth examining both the curriculum as written and the conditions under which educators put it into practice. UNESCO’s Reporting on Artificial Intelligence: A Handbook for Journalism Educators (2023), edited by Maarit Jaakkola, offers a further reference point: it treats AI not only as a tool students must learn to use, but as a social issue they must learn to report on. Tejedor et al. · Wenger et al. · UNESCO handbook

What is new is the connection. Foresight studies imagine futures for journalism but rarely ask what those futures would demand of graduates; curriculum studies describe what is taught but rarely test it against possible futures. The UNESCO Chair’s programme brings the two together in a single comparative design, and adds a third element: the expectations and collective climates of hope, anxiety or resignation that shape whether educators and students are ready to change.

Where media and information literacy fits

Media and information literacy, which includes AI literacy, does not replace the analysis of the sector or of expectations; it connects them. Journalists will need to evaluate AI systems as tools, investigate the institutions that develop and deploy them, and explain their uses and limitations to the public. They can also make their own reporting easier to scrutinise by attributing sources clearly, explaining how evidence was verified and making corrections visible. These practices support the public’s ability to assess claims, seek reliable information and participate in democratic debate. They connect the programme with UNESCO’s Recommendation on the Ethics of Artificial Intelligence and its new companion guide to the Media and Information Literacy Curriculum, AI Answers. You Decide. UNESCO Recommendation · UNESCO curriculum guide

The UNESCO Chair is assembling the programme’s research teams. Planned outputs include a monitor of the three tensions and scenarios for journalism up to 2036 built from it; a study of the expectations of journalists, educators and students; an open international map of undergraduate and master’s programmes; a comparative report on the gap between current teaching and anticipated needs; and a media and information literacy and AI literacy competence framework for journalism education, tested against the scenarios and against demand from the sector. Universities, research groups, educators and media professionals can express interest through the MILID Network, whether they wish to coordinate a country or regional team, contribute to curriculum mapping, or take part in expert consultations and scenario workshops. Download the prgramme.