Synthetic Realities: Understanding AI-Generated Visual Vaccine Misinformation in Digital Public Discourse - A Feasibility Study
Lead Research Organisation:
MANCHESTER METROPOLITAN UNIVERSITY
Abstract
Artificial intelligence can now generate images, videos, and text indistinguishable from authentic content. While offering exciting creative possibilities, this technology poses unprecedented challenges to public discourse and social trust. AI-generated health misinformation threatens public health, particularly vaccine confidence. This fellowship develops systematic methods for understanding how synthetic visual content about vaccines and health interventions spreads across social media platforms, exploiting health anxieties and undermining evidence-based medicine.
This fellowship makes two critical contributions to digital society research through a feasibility study of synthetic visual content analysis. First, it advances understanding of AI-generated fakeness, platform governance challenges, and democratic discourse transformation in the synthetic media era. Second, it develops Synthetic Discourse Analysis (SDA) as a methodological innovation combining advanced social listening technologies with qualitative discourse analysis. Rather than treating AI-generated content as purely technical detection, this dual approach reveals how synthetic content operates as both computational phenomenon and social process, fundamentally reshaping community meaning-making.
Building on platform studies and digital politics scholarship, the research maps how AI-generated misinformation spreads across Twitter/X, Facebook, BlueSky, Instagram, and TikTok, identifying amplification pathways and audience targeting strategies. Using cutting-edge social listening tools and custom analysis frameworks, the research analyses social media discourse to understand how synthetic content integrates with organic user discourse and influences public conversation.
A crucial focus is understanding how AI-generated misinformation disproportionately affects marginalised communities, who often lack digital literacy resources for fact-checking health misinformation, yet bear the greatest consequences of misinformation campaigns, including lower vaccine uptake in deprived areas with high measles outbreak rates. Addressing this involves examining targeting patterns and identifying communities using platforms beyond conventional monitoring systems. Community consultation ensures methodological approaches serve social justice by centring affected voices in intervention design.
This fellowship advances multiple fields whilst developing practical interventions. It extends digital society research by revealing AI's transformation of public discourse, contributes to platform governance studies through synthetic content analysis, and develops SDA methodology as a reproducible framework. These provide tools for researchers, journalists, and civil society whilst contributing evidence to the SDR UK Data Access Taskforce.
Building on expertise in Visual AI Literacy education and social media discourse analysis, the research produces community-based detection methods, digital literacy resources, and policy recommendations. The open-source toolkit enables researchers globally to apply these approaches to different contexts and misinformation domains.
Academic impact encompasses advancing understanding of synthetic content's social integration whilst developing methodological frameworks for social media research. Policy impact includes direct contribution to UK regulatory frameworks through evidence provision for Online Safety Act implementation and Data Use and Access Act 2025 development.
The UK faces particular vulnerabilities to AI-generated misinformation, demonstrated during Brexit debates, Covid-19 and NHS misinformation campaigns, and recent election cycles. This research provides crucial evidence for implementing robust regulatory responses whilst ensuring protection extends equitably across communities.
This fellowship contributes to Smart Data Research UK's digital society priorities through theoretical advancement and methodological innovation. Combining substantive insights into AI's societal impacts with innovative analytical methods and community-centred research ethics ensures rapid technological development is accompanied by rigorous attention to social consequences and democratic values.
Expected outcomes include high-impact academic publications, policy briefings for UK government departments, training programmes for journalists and educators, and accessible digital literacy resources for communities most affected by AI-generated misinformation.
This fellowship makes two critical contributions to digital society research through a feasibility study of synthetic visual content analysis. First, it advances understanding of AI-generated fakeness, platform governance challenges, and democratic discourse transformation in the synthetic media era. Second, it develops Synthetic Discourse Analysis (SDA) as a methodological innovation combining advanced social listening technologies with qualitative discourse analysis. Rather than treating AI-generated content as purely technical detection, this dual approach reveals how synthetic content operates as both computational phenomenon and social process, fundamentally reshaping community meaning-making.
Building on platform studies and digital politics scholarship, the research maps how AI-generated misinformation spreads across Twitter/X, Facebook, BlueSky, Instagram, and TikTok, identifying amplification pathways and audience targeting strategies. Using cutting-edge social listening tools and custom analysis frameworks, the research analyses social media discourse to understand how synthetic content integrates with organic user discourse and influences public conversation.
A crucial focus is understanding how AI-generated misinformation disproportionately affects marginalised communities, who often lack digital literacy resources for fact-checking health misinformation, yet bear the greatest consequences of misinformation campaigns, including lower vaccine uptake in deprived areas with high measles outbreak rates. Addressing this involves examining targeting patterns and identifying communities using platforms beyond conventional monitoring systems. Community consultation ensures methodological approaches serve social justice by centring affected voices in intervention design.
This fellowship advances multiple fields whilst developing practical interventions. It extends digital society research by revealing AI's transformation of public discourse, contributes to platform governance studies through synthetic content analysis, and develops SDA methodology as a reproducible framework. These provide tools for researchers, journalists, and civil society whilst contributing evidence to the SDR UK Data Access Taskforce.
Building on expertise in Visual AI Literacy education and social media discourse analysis, the research produces community-based detection methods, digital literacy resources, and policy recommendations. The open-source toolkit enables researchers globally to apply these approaches to different contexts and misinformation domains.
Academic impact encompasses advancing understanding of synthetic content's social integration whilst developing methodological frameworks for social media research. Policy impact includes direct contribution to UK regulatory frameworks through evidence provision for Online Safety Act implementation and Data Use and Access Act 2025 development.
The UK faces particular vulnerabilities to AI-generated misinformation, demonstrated during Brexit debates, Covid-19 and NHS misinformation campaigns, and recent election cycles. This research provides crucial evidence for implementing robust regulatory responses whilst ensuring protection extends equitably across communities.
This fellowship contributes to Smart Data Research UK's digital society priorities through theoretical advancement and methodological innovation. Combining substantive insights into AI's societal impacts with innovative analytical methods and community-centred research ethics ensures rapid technological development is accompanied by rigorous attention to social consequences and democratic values.
Expected outcomes include high-impact academic publications, policy briefings for UK government departments, training programmes for journalists and educators, and accessible digital literacy resources for communities most affected by AI-generated misinformation.
People |
ORCID iD |
| Samantha Martin (Principal Investigator / Fellow) |