AI-driven microtargeting exploits democratic vulnerabilities as EU regulations fail to address algorithmic manipulation of electoral processes

Indigenous epistemologies reject the hyper-individualism of AI microtargeting, instead emphasizing communal decision-making and relational accountability. Traditional knowledge systems, such as those in Māori or Quechua cultures, frame data as a collective resource rather than a commodity, directly challenging the extractive logic of surveillance capitalism. However, these perspectives are systematically excluded from EU regulatory debates, which treat data as a purely technical and economic asset.

The use of AI in elections mirrors historical patterns of propaganda and psychological manipulation, from Nazi-era radio broadcasts to CIA-funded disinformation during the Cold War. The EU’s regulatory failure echoes the 19th-century laissez-faire approach to industrial capitalism, where unchecked corporate power led to social upheaval. The study’s focus on 'current legislation' ignores the long-standing co-optation of democratic institutions by corporate interests, as seen in the revolving door between Silicon Valley and EU policymakers.

Non-Western democracies have grappled with AI-driven electoral interference for over a decade, with Brazil’s 2018 WhatsApp scandal serving as a cautionary tale for the EU. In India, the Aadhaar biometric system has been weaponized to suppress voter turnout among marginalized castes, while in Kenya, AI-powered disinformation fueled ethnic violence in 2007–2008. These cases reveal a global pattern where platform monopolies and state actors collaborate to manipulate electoral outcomes, often with impunity.

The study’s findings align with peer-reviewed research on algorithmic persuasion, such as the 2020 *PNAS* paper demonstrating how Facebook’s ad-targeting algorithms can shift voting behavior by 3–5%. However, the research underplays the role of reinforcement learning in optimizing microtargeting, which evolves faster than regulatory frameworks. The EU AI Act’s risk-based approach fails to account for the emergent properties of AI systems, where small changes in algorithmic parameters can have outsized effects on voter behavior.

Artistic movements like *détournement* and *culture jamming* have long exposed the manipulative power of media, from Guy Debord’s *Society of the Spectacle* to Banksy’s subversive street art. Indigenous storytelling traditions, such as the Māori *pūrākau*, frame truth as a communal and spiritual process, directly opposing the atomized, data-driven narratives of AI microtargeting. However, these creative and spiritual critiques are marginalized in favor of technocratic solutions that treat symptoms rather than root causes.

Scenario modeling suggests that without structural reforms, AI-driven electoral interference could escalate into a 'post-democratic' era where outcomes are predetermined by algorithmic control. The EU’s current approach risks normalizing surveillance capitalism in governance, while alternative models—such as citizen data trusts or algorithmic transparency mandates—remain unexplored. Future-proofing democracy requires preemptive regulation of platform monopolies and the integration of participatory design principles into electoral systems.

Marginalized communities—particularly racialized groups, low-income voters, and non-Western populations—are disproportionately targeted by AI microtargeting due to data poverty and weak privacy protections. The study’s focus on 'voter behavior' obscures the fact that these groups are often disenfranchised rather than 'influenced,' as their data is used to suppress turnout or spread disinformation. Grassroots organizations like the *Digital Freedom Fund* and *Access Now* have documented how AI exacerbates existing inequalities, yet their insights are excluded from policy discussions.