This debate asks whether a few huge, global AI models will effectively crowd out smaller, local-language or domain-specific systems. One side may claim that scale, resources, and data give global models an unbeatable advantage, which could marginalize smaller languages and local ecosystems. The other side will highlight the strengths of local models: cultural nuance, legal and educational control, data sovereignty, and the ability to tailor systems closely to community needs. Underneath is a question about digital power: centralization vs. diversity and local autonomy.
Agree Position
Disagree Position
Senior Research Associate at the Harvard Labour and Worklife Program and Vice-President of Kozminski University for AI, heading the Human-Machine Interaction Research Center. Completed post-doctoral research at MIT's Center for Collective Intelligence in Boston. Co-author of "Collaborative Society" (The MIT Press) and "Strategizing AI in Business and Education" (Cambridge University Press) with Dariusz Jemielniak.
Co-founder and chief architect at Azurro S.C., where he helps clients from around the world solve problems related to data processing and analysis. Together with his teams, he designs, builds, and supports the maintenance and development of solutions in the areas of Machine Learning and Big Data. In Poland, he works mainly with companies in the media industry, providing automatic transcription, analysis, and indexing services for audiovisual materials.
