Thinking with AI: How Gen AI is re-shaping political autonomy and human judgement

Most conversations about ChatGPT and Generative AI focus on dramatic risks: fake news, automated propaganda, manipulated voters, or AI-generated political content. But a quieter—and perhaps more important—question is often overlooked: what happens when ordinary people use ChatGPT to understand political events, build arguments, make sense of complex issues, or decide what sounds convincing?

This talk presents research suggesting that the greatest risk is not only that AI might deceive us from the outside. It is also that we may gradually hand over parts of our own judgement, language, and imagination because these systems are fast, confident, and incredibly useful.

Drawing on a research project with university students in Spain and Germany, participants used ChatGPT-like tools during simulated political crises, taking on the roles of political parties, public institutions, civil society groups, and protest movements while making decisions under time pressure.

The findings reveal that the influence of AI does not begin inside the chatbot. Instead, it depends on the freedom people already have before they start using it: enough time to think, opportunities for genuine dialogue, space for disagreement, and the confidence to develop their own ideas before asking the machine for help. Without these conditions, ChatGPT can quietly reshape the way we think, speak, and judge. With them, it becomes a powerful tool that supports human thinking rather than replacing it.

Rather than asking whether AI is simply good or bad, this talk invites the audience to consider a more important question: when we use ChatGPT to think about politics, are we strengthening our own judgement—or slowly giving away the conditions that make independent judgement possible?

What we will cover

• Why the political impact of ChatGPT goes far beyond fake news, propaganda, or election manipulation.
• How people use Generative AI to understand events, build arguments, simplify complex issues, and decide what sounds persuasive.
• What simulations of political crises in Spain and Germany reveal about judgement, decision-making, and time pressure.
• Why the effects of AI depend less on writing better prompts and more on having time, dialogue, disagreement, creativity, and autonomy.
• How citizens, educators, journalists, and political organisations can protect human judgement by creating spaces for critical thinking before turning to AI. 

The Speaker: Dr. Marco Guglielmo

Marie Curie Postdoctoral Fellow at the University of Valencia, specializing in political communication, digital politics and Generative AI. His research explores how digital platforms and AI are reshaping political discourse and the way people engage with politics. He is the author of The Left and Digital Politics: Political Parties from Platform Neoliberalism to Platform Socialism and has published in leading journals including Government & Opposition and the International Journal of Communication.