Digital Researcher Blogs

Discover blogs from our talented researcher community here at The University of Manchester.

October 2026

GenAI and Politics: The Fifth Age of Political Campaigning?

Conor Gaughan - Research Fellow in AI Trust and Security at the Centre for Digital Trust and Security   

When the World Wide Web launched in the 1990s, it fundamentally altered the nature of political campaigning. Politicians were suddenly afforded a whole new range of ways to communicate with their constituents like setting up webpages, writing weblogs, sending emails, and circulating newsletters. What started off as a relatively static and one-dimensional form of communication eventually morphed into what scholars now call the “fourth age” of political campaigning (Blumler, 2016); an age defined by multinetwork communication, data-driven microtargeting, algorithmic curation, and a permanently online electorate. 

The public launch of ChatGPT in late 2022 sparked a new juncture in this timeline, leading to the proliferation of generative AI tools like Gemini, Claude and Copilot. These tools can generate text, images, videos, audio and other forms of digital media at scale with a simple written prompt. They have dramatically improved in their sophistication over the last four years and there are increasing concerns about their potential to destabilise the democratic process. Recent high-profile allegations of generative AI use by politicians to write parliamentary speeches, compile evidentiary reports, respond to constituent emails, and create artificial chatbots of themselves has garnered considerable criticism (James, 2025; Kelly and Sommerlad, 2025; Moss and Vincent, 2025).  

While there was no evidence to suggest that generative AI meaningfully impacted the results of the UK or European elections in 2024, it showed its ability to undermine trust in genuine online sources, incite hatred against political candidates, and incentivise unethical election behaviour by political parties (Stockwell, 2024). Yet, due to the relative infancy of generative AI, our knowledge of exactly how it might be transforming political communication is extremely limited. This leaves a critical knowledge gap which weakens our ability for effective regulation and governance of these tools within the electoral process.  

My research seeks to fill this gap over the next three years by building a comprehensive portfolio of evidence about how the use of generative AI in political communication is affecting public trust. To do so, my work will employ a range of methods including the detection of AI-mediated content by politicians in parliament and online, experiments to test the impact of its use on public trust, and interviews with MPs to better understand their side of the story.  

Research tells us that UK voters are broadly accepting of MPs using AI tools to decrease their workload, but they demand transparency, accountability and, above all, human control (Jarvis, 2025). There is a concern that by relying too heavily on these tools in their day-to-day activities, MPs may be outsourcing their representative duties to artificial intelligence. In a period of major uncertainty, where political trust is flatlining and misinformation is rife, it is crucial that elected officials are made aware of the risks of using AI in their communication, what their voters expect from them if they use it, and that they are transparent and accountable when they do.

REFERENCES 

Blumler, J.G., (2016). “The fourth age of political communication”, Politiques de communication, 6(1), pp.19-30. doi: https://doi.org/10.3917/pdc.006.0019  

James, R., (2025). Labour MPs accused of using ChatGPT to write speeches. The Independent. Available at: https://www.independent.co.uk/news/uk/home-news/labour-mp-ai-chat-gpt-speeches-tugendhat-b2823339.html [accessed 20/01/26] 

Jarvis, H. (2025). Public draws red line on AI in parliament. Brunel: University of London. Available at https://www.brunel.ac.uk/news-and-events/news/articles/Public-draws-red-line-on-AI-in-Parliament [accessed 26/01/2026] 

Kelly, H. and Sommerlad, N. (2025). MP spotted using ChatGPT to respond to constituents on the train. The Mirror. Available at: https://www.mirror.co.uk/news/politics/mp-spotted-using-chatgtp-respond-35752561 [accessed 20/01/26] 

Moss, A., and Vincent, J., (2025). MP becomes first to create himself as an AI bot. BBC News. Available at: https://www.bbc.co.uk/news/articles/cy5pr3q6lrpo [accessed 20/01/26] 

Stockwell, S., (2024). AI-Enabled Influence Operations: Threat Analysis of the 2024 UK and European Elections. Centre for Emerging Technology and Security. Available at: https://cetas.turing.ac.uk/publications/ai-enabled-influence-operations-threat-analysis-2024-uk-and-european-elections [accessed 28/01/26]