Technological Forecasting and Social Change—AI for whom and by whom?

AI for whom and by whom? Cultural bias and the institutional and social shaping of large language models

Guest editors:

• Beatrice Orlando (Lead Guest Editor), University of Ferrara, Italy, Beatrice.orlando@unife.it
• Marcus Wagner, University of Augsburg, Germany. marcus.wagner@uni-a.de
• Janet Rafner, University of Southern Denmark, Denmark. jraf@sam.sdu.dk
• Thierry Burger-Helmchen, University of Strasbourg, BETA, France. burger@unistra.fr

Special issue information:

Rationale and Scope

Large language models (LLMs) are rapidly becoming embedded in decision making and other processes across governments, corporations, and civil society (Dwivedi et al., 2021; Haefner et al., 2021). These systems promise efficiency gains, new forms of knowledge production, and long term technological change. However, the development and training of most current LLMs draw on datasets and take place within innovation ecosystems that are deeply rooted in Western cultural norms, English-language online text collections, and the problem-solving logics of a small number of global technology hubs (Tao et al., 2024). These ecosystems matter because they concentrate data resources, funding, and expertise in specific regions, thereby shaping the questions that are prioritized, the forms of knowledge that are encoded, and the applications for which these models are primarily designed. As a result, they tend to reflect problem definitions and solution patterns that are closely tied to specific countries or regional blocs, which are then applied to ostensibly global contexts, so that in practice many LLMs function less as universal infrastructures and more as regionally shaped tools (Harari, 2024). Recent scholarship has increasingly highlighted this situation, emphasizing that emerging technologies are not neutral.

A parallel strand of work has begun to conceptualize AI as an emerging general method of invention that transforms scientific discovery, innovation trajectories, and the organization of research systems (Bianchini et al., 2022). LLMs play an increasing role in these transformations, as tools for literature reviews, hypothesis generation, coding, and research communication. Yet the governance of LLMs remains uneven, fragmented across jurisdictions, and often focused on abstract principles rather than on the concrete distribution of capabilities and benefits across actors, sectors, and regions (Mensah and van Wynsberghe, 2025, Orlando et al., 2022). Recent reviews of AI related work indicate that, while attention to innovation management and sustainability implications is growing, there is still limited focus on cultural pluralism, institutional diversity, and the geopolitical structuring of AI development (Di Vaio et al., 2021; Natale and Ballatore, 2020). There is even less systematic attention to how LLMs are shaped by, and in turn reshape, national and regional innovation systems, policy frameworks, and knowledge cultures.

As LLMs become central to research and strategic decision-making, their embedded cultural biases have material consequences for whose knowledge is legitimized, whose perspectives are marginalized, and which futures are rendered thinkable (He et al., 2025; He and Burger-Helmchen, 2025). From a social impact perspective, culturally biased LLMs can reinforce digital colonialism, knowledge extractivism, and linguistic marginalization, particularly in the Global South (Böhmecke-Schwafert and García Moreno, 2023; Parthasarathy et al., 2021). They may skew policy priorities toward Western techno managerial solutions, while obscuring locally grounded ecological practices, community based governance, and indigenous innovation traditions.

Crucially, LLMs are not developed or deployed in a homogeneous way worldwide. Distinct ecosystems and networks of enablement, production, and use are emerging, structured around powerful cloud providers, platform firms, data holders, and sectoral complementors. Building on research that maps the evolutionary dynamics of the AI ecosystem (Jacobides et al., 2021), we can distinguish between actors that provide enabling infrastructures (compute, data, foundational LLMs), those that produce applications and services on top of these models, and those that integrate LLMs into domain specific practices in fields such as mobility, energy, health, finance, or public administration. These ecosystems differ across countries and regional blocs, with contrasting configurations in North America, Europe, China, and the Global South in terms of industry architecture, public policy, scientific capacity, and patterns of collaboration between firms, states, and research institutions. This raises core questions about who builds LLMs, for which applications, on whose infrastructure, and with what implications for inclusion, and global power asymmetries.

Public research organizations, and in particular universities and university networks, have long been key actors in technological development and foundational research, providing mission oriented innovation capacity, early stage knowledge generation, and high risk exploratory work that the private sector is often unwilling to finance (Mazzucato and Semieniuk, 2017). Yet their specific contribution to LLM development frequently remains under acknowledged, even though universities provide much of the basic research in machine learning, a significant share of open models and datasets, and the training of highly skilled researchers who subsequently shape industrial and policy agendas around AI (Guo et al., 2026; Mangiò et al., 2025; Park et al., 2026). Moreover, patterns of collaboration between universities, firms, and public agencies vary substantially across countries and regional blocs, with differences in funding regimes, data access, and governance arrangements influencing which LLM research agendas are pursued, how risks and benefits are distributed, and the extent to which local languages and knowledge systems are incorporated into model development (Alita, 2025; Li et al., 2026). This raises further questions about how European university-industry collaborations in AI and LLMs contribute to or counterbalance the dominance of a few large corporate actors, the extent to which university alliances and consortia can pool limited computing resources, data, and expertise to build regionally grounded LLMs, and how national and European-level AI initiatives allocate roles and resources between public research organizations and private firms.

Against this background, the Special Issue titled “AI for whom and by whom? Cultural bias and the institutional and social shaping of large language models” asks explicitly who is developing LLMs, for what purposes, in which organizational, institutional, and territorial contexts. It invites analyses that treat LLMs not only as technical artifacts, but as outcomes of specific institutional settings, innovation policies, business models, and knowledge cultures, embedded in concrete ecosystems and networks.

Key themes and types of contribution

This Special Issue welcomes empirical papers drawing on all type of methods. Contributions need to engage with LLMs and their institutional and cultural implications, and should aim to shed light on one or more of the following, or closely related, themes:

• How LLM training data, model design, and deployment practices are embedded in specific cultural, linguistic, institutional, and regional contexts, and how this shapes what is represented or silenced.
• Cultural bias, digital colonialism, knowledge extractivism, and the treatment of local languages, indigenous knowledge, and alternative epistemologies in LLM development and use.
• The structure and evolution of LLM related ecosystems and value chains, including the roles and power of cloud providers, platform firms, data holders, start-ups, and sector specific complementors.
• Comparative analyses of national or regional LLM ecosystems, examining differences in industry architecture, public policy, data governance, and strategic positioning across countries or blocs such as the United States, the European Union, China, and regions of the Global South.
• The contributions and strategies of public research organizations, universities, and university networks in LLM research and deployment, including university industry collaborations, national initiatives, and cross border alliances formed to pool scarce compute, data, and expertise.
• Case studies exploring how participatory foresight, co-design, or public deliberation can shape LLM development and deployment toward more socially accountable, sustainable and future-oriented outcomes.

Manuscript submission information:

The Special Issue will be open for submission on September 1, 2026. Please submit your manuscript to Editorial Manager® before the submission deadline April 30, 2027, under article type name "VSI: AI for whom by whom".

For inquiries or informal discussions, please contact: beatrice.orlando@unife.it

References:

Alita, L., 2025. Enhancing citizen engagement in urban greening: The potential of large language models in value co-creation. Technological Forecasting and Social Change 216, 124134. https://doi.org/10.1016/j.techfore.2025.124134

Bianchini, S., Müller, M., Pelletier, P., 2022. Artificial intelligence in science: An emerging general method of invention. Research Policy, 51, 104604. https://doi.org/10.1016/j.respol.2022.104604

Böhmecke-Schwafert, M., García Moreno, E., 2023. Exploring blockchain-based innovations for economic and sustainable development in the global south: A mixed-method approach based on web mining and topic modeling. Technological Forecasting and Social Change 191, 122446. https://doi.org/10.1016/j.techfore.2023.122446

Di Vaio, A., Palladino, R., Pezzi, A., Kalisz, D.E., 2021. The role of digital innovation in knowledge management systems: A systematic literature review. Journal of Business Research 123, 220–231. https://doi.org/10.1016/j.jbusres.2020.09.042

Dwivedi, Y.K., Hughes, L., Ismagilova, E., Aarts, G., Coombs, C., Crick, T., Duan, Y., Dwivedi, R., Edwards, J., Eirug, A., Galanos, V., Ilavarasan, P.V., Janssen, M., Jones, P., Kar, A.K., Kizgin, H., Kronemann, B., Lal, B., Lucini, B., Medaglia, R., Le Meunier-FitzHugh, K., Le Meunier-FitzHugh, L.C., Misra, S., Mogaji, E., Sharma, S.K., Singh, J.B., Raghavan, V., Raman, R., Rana, N.P., Samothrakis, S., Spencer, J., Tamilmani, K., Tubadji, A., Walton, P., Williams, M.D., 2021. Artificial Intelligence (AI): Multidisciplinary perspectives on emerging challenges, opportunities, and agenda for research, practice and policy. International Journal of Information Management 57, 101994. https://doi.org/10.1016/j.ijinfomgt.2019.08.002

Guo, Z., Yang, G., Wu, W., 2026. Automatic synthesis of econometric empirical research results using large language model: A case study of digitalization-greening relationships. Technological Forecasting and Social Change 222, 124370. https://doi.org/10.1016/j.techfore.2025.124370

Haefner, N., Wincent, J., Parida, V., Gassmann, O., 2021. Artificial intelligence and innovation management: A review, framework, and research agenda. Technological Forecasting and Social Change 162, 120392. https://doi.org/10.1016/j.techfore.2020.120392

Harari, Y.N., 2024. Nexus: A Brief History of Information Networks from the Stone Age to AI. Penguin Random House, New York.

He, X., Antonczak, L., Burger-Helmchen, T., 2025. Revisiting ba: How generative AI transforms knowledge creation. Journal of Knowledge Management, https://doi.org/10.1108/JKM-07-2025-1065

He; X., Burger-Helmchen, T., 2025, Evolving Knowledge Management: Artificial Intelligence and the Dynamics of Social Interactions, IEEE Engineering Management Review, 53, 5, 215-231, https://doi.org/10.1109/EMR.2024.3519342

Jacobides, M.G., Brusoni, S., Candelon, F., 2021. The Evolutionary Dynamics of the Artificial Intelligence Ecosystem. Strategy Science 6, 412–435. https://doi.org/10.1287/stsc.2021.0148

Li, Y., Cao, D., Hewitt, M., Shan, S., 2026. Using large-language models to analyse social media engagement with the public sector communication on circular economy: Evidence from X. Technological Forecasting and Social Change 222, 124400. https://doi.org/10.1016/j.techfore.2025.124400

Mangiò, F., Pedeliento, G., Wassler, P., Williams, N., 2025. Discursively negotiating AI: A social representation theory approach to LLM-based chatbots. Technological Forecasting and Social Change 221, 124352. https://doi.org/10.1016/j.techfore.2025.124352

Mazzucato, M., Semieniuk, G., 2017. Public financing of innovation: new questions. Oxford Review of Economic Policy 33, 24–48. https://doi.org/10.1093/oxrep/grw036

Mensah, M., van Wynsberghe, A. Sustainable AI meets feminist African ethics. AI and Ethics, 5, 4293–4303 (2025). https://doi.org/10.1007/s43681-025-00705-9

Natale, S., Ballatore, A., 2020. Imagining the thinking machine: Technological myths and the rise of artificial intelligence. Convergence 26, 3–18. https://doi.org/10.1177/1354856517715164

Orlando, B., Ballestra, L.V., Scuotto, V., Pironti, M., Giudice, M.D., 2022. The Impact of R&D Investments on Eco-Innovation: A Cross-Cultural Perspective of Green Technology Management. IEEE Transactions on Engineering Management 69, 2275–2284. https://doi.org/10.1109/TEM.2020.3005525

Park, S., Kim, G., Lee, S., 2026. Evaluating the value of LLMs in patent-based technology intelligence: Toward increasing efficiency and reducing expert dependency. Technological Forecasting and Social Change 222, 124375. https://doi.org/10.1016/j.techfore.2025.124375

Parthasarathy, B., Dey, S., Gupta, P., 2021. Overcoming wicked problems and institutional voids for social innovation: University-NGO partnerships in the Global South. Technological Forecasting and Social Change 173, 121104. https://doi.org/10.1016/j.techfore.2021.121104

Tao, Y., Viberg, O., Baker, R. S., & Kizilcec, R. F. (2024). Cultural bias and cultural alignment of large language models, PNAS Nexus, Volume 3, Issue 9, September 2024, p.346.

Keywords:

Large Language Models (LLMs); Generative AI Governance and Regulation; Cultural and Linguistic Bias in AI; Global AI Innovation Ecosystems; Institutional and Geopolitical Dynamics of AI; Digital Colonialism in Artificial Intelligence