With early voting already underway across the country, a growing number of Americans are experimenting with a new tool as they prepare to cast their ballots: artificial intelligence chatbots. According to reporting from NPR, some voters are now asking AI systems to help them research races, fact-check claims, and even weigh in on which candidates best align with their personal values. The trend marks a notable shift in how citizens approach civic decision-making, raising fresh questions about accuracy, bias, and the role of technology in democratic processes.
One example comes from Morgantown, West Virginia, where 40-year-old Adam Johnson sat on his porch one September evening to review his ballot for the upcoming midterm election. Rather than turning to a friend or family member for discussion, Johnson spent hours in conversation with ChatGPT, the chatbot developed by OpenAI. Together, they reviewed the various races and candidates on his ballot, with Johnson working to determine who best represented his values, according to NPR.
Johnson, a graphic designer who is studying to become an illustrator, described ChatGPT as a significant presence in his daily life. He relies on it for tasks ranging from meal planning to job hunting and other household needs. When it came to the election, he first asked the chatbot to compile a list of all the races appearing on his ballot, then engaged in a deeper discussion about the candidates.
Interestingly, Johnson noted that in a previous conversation, the AI had appeared reluctant to engage with political topics. He recalled that the chatbot had initially responded with a simple refusal when he tried to discuss politics. That hesitancy, however, did not prevent him from pursuing the conversation further in this instance.
The NPR report, which aired on Weekend Edition Sunday and was written by Maham Javaid, highlights that voters are increasingly turning to AI chatbots for fact-checking and side-by-side comparisons of candidates. This behavior reflects a broader integration of AI tools into everyday life, where people already use them for everything from drafting emails to planning meals. The midterm elections appear to be the latest arena for this technological experimentation.
While the full implications of this trend remain unclear, it raises several concerns. AI chatbots are powered by large language models that can generate confident-sounding but inaccurate information, a phenomenon often called hallucination. When applied to elections, such errors could mislead voters about candidate positions, voting deadlines, or polling locations. The lack of transparency in how these systems weigh sources also makes it difficult to assess potential political biases.
Beyond accuracy, there is the question of privacy. Voters who share their political leanings, locations, or concerns with a chatbot may not fully understand how that data is stored or used. Unlike conversations with a neighbor or a trusted friend, interactions with AI systems are logged and processed by private companies, which may have their own commercial or policy interests.
Supporters of the practice argue that AI can help voters navigate an increasingly complex information environment. With countless races, propositions, and candidates on a typical ballot, having a tool that can summarize and compare positions could be a valuable time-saver. For voters who lack easy access to nonpartisan voter guides or who feel overwhelmed by the volume of information, a chatbot might seem like a convenient alternative.
Still, experts have cautioned that AI systems are not neutral arbiters. They are trained on vast datasets that may contain outdated or skewed information, and their responses can vary depending on how questions are phrased. A voter asking "Who should I vote for?" may receive a very different answer than one asking "What are the main differences between these two candidates?" The former invites a normative judgment, while the latter is more factual — though still subject to error.
The NPR story also underscores how quickly AI has moved from novelty to utility in civic life. Just a few years ago, the idea of consulting a chatbot about one's ballot would have seemed futuristic. Now, as the midterms proceed, it is becoming a reality for some voters. Whether this leads to more informed or more confused electorates remains an open question, but the trend is unlikely to fade as AI tools become more capable and more deeply embedded in daily routines.
Election officials and technology watchdogs may soon need to address this phenomenon. Some jurisdictions already publish official voter guides, and nonpartisan organizations offer candidate comparison tools. If AI chatbots become a primary source of election information for a meaningful share of the population, the integrity of those tools — and the transparency of the companies behind them — will face heightened scrutiny.
For now, the story of Adam Johnson and others like him illustrates a quiet but significant change in how democracy is practiced. The act of voting has always been shaped by the information available to citizens, from pamphlets and newspapers to television and social media. AI chatbots are simply the latest chapter in that evolution — one that blends convenience with risk, and personalization with the potential for manipulation.
As early voting continues and Election Day approaches, observers will be watching to see whether this experiment in AI-assisted voting spreads further. The outcome may influence not only who wins the midterms but also how future generations think about the intersection of technology and civic duty.
(0)