Beyond the Hype: How We’re Overstating AI’s Impact on Elections
With elections happening across the globe this year, the air is thick with speculation about how AI might influence our democratic processes. Predictions paint a picture of doom, warning of deepfakes that could trick voters, AI-driven misinformation campaigns, and hyper-targeted ads that might sway entire populations. But as more elections come and go, a clearer picture is emerging—one that challenges many of these fears. Far from ushering in an era of AI-dominated election tampering, the evidence so far suggests that AI's impact on voting behaviours may be much less substantial than initially feared.
Why? It turns out that predicting the end of democracy via AI is far simpler than the reality, which reveals deep-rooted factors shaping voter behaviour that AI alone has limited power to disrupt.
The Reality Behind AI’s Limited Influence
For any technology, AI included, to truly change the outcomes of elections, it would need to overcome some very human obstacles: the way we process information, the context in which we receive it, and our underlying values. Let’s explore why these factors significantly limit AI’s influence.
1. Persuasion Isn’t as Easy as It Looks
Research has long shown that shifting people’s deeply held beliefs and values is incredibly challenging. Most of us update our beliefs only slightly, even when presented with new information, and that update rarely drives us to act differently. Campaigns pour millions into ads aimed at convincing us to vote one way or another, but study after study has found that these efforts typically yield minor results at best. So, while AI-generated content might look threatening on the surface, it runs up against the same limits: we’re simply not that easily swayed.
2. Cutting Through the Noise
Even the most sophisticated AI-driven campaigns must still contend with the sheer volume of information online. Our news feeds, social platforms, and inboxes are packed with messages from all sides. For AI to actually influence voter behaviour, it has to get noticed—and not just noticed but trusted, remembered, and acted upon. With an overwhelming flow of content, AI-generated messages often get lost in the noise, making it hard for any one piece of content to stand out or create lasting effects.
3. The Downside of Personalisation
AI may be able to tailor messages to individual voters, but hyper-targeting can sometimes backfire. People increasingly notice when they’re being “over-targeted” with content, and this can lead to annoyance or distrust. Many of us have experienced the “creep factor” when ads seem a little too tailored, making us less receptive to the message. Recent studies suggest that voters aren’t just unimpressed by AI-targeted political ads; they’re actively turned off by them. Campaign strategists appear to recognize this, as they’re using AI mostly for behind-the-scenes operations, such as managing outreach or fundraising, rather than attempting to craft persuasive voter messaging.
4. Voting Is More Than Information
Voting decisions are influenced by a complex web of factors—gender, class, values, identities, and lifelong socialization processes, to name a few. Information, especially if it contradicts deeply held beliefs or values, is only a small piece of the puzzle. This explains why AI-generated messages, even if carefully targeted, rarely change votes on their own. Voting behaviour is deeply entrenched, and while AI might make messaging a bit more efficient, it’s unlikely to reshape fundamental political beliefs or identities.
Why Fixating on AI Misses the Point
The focus on AI as a singular threat to elections risks overshadowing other issues that have far more immediate impacts on democracy. It’s essential not to lose sight of the longstanding problems that threaten democratic systems, like mass voter disenfranchisement, attacks on journalists, threats to free speech, and political misinformation not involving AI. Many nations facing these issues this year are hybrid or authoritarian regimes, where elections are often symbolic gestures rather than genuine opportunities for change. For citizens in such contexts, AI-generated misinformation is far from their most pressing concern.
The Real Impact of Overstating AI’s Role in Elections
When we overemphasize AI as the chief threat, we inadvertently fuel public fear and confusion. This leads to a decline in trust—not just in technology but in the institutions that use and regulate it. Constant warnings about AI-driven election interference risk creating a sense of inevitable manipulation, causing people to lose faith in democratic processes and even in the legitimacy of election outcomes.
Meanwhile, the obsession with AI lets politicians off the hook. Blaming technology can be a convenient way to sidestep much-needed reforms. For some leaders, it’s easier to point to AI than to fix voting infrastructure or address systemic inequalities.
Shifting Focus Back to Democracy’s Core Challenges
AI will undoubtedly play some role in the future of elections. But rather than viewing it as an existential threat, we’d do better to see it as a tool that must be handled with care, within the broader context of democratic resilience. Efforts to safeguard democracy are better spent focusing on accountability, transparency, and robust democratic institutions. Remember these aren’t abstract ideals—they’re practical foundations that can withstand the tests of new technologies and the challenges to come.
By bringing our attention back to the structural and social factors that shape elections, we’re more likely to address the real, underlying threats to democracy—leaving AI as just one part of a much larger story.
