The New Battlefield of Election Misinformation: How False Claims Are Reshaping Immigrant Communities

The spread of election misinformation has entered a dangerous new phase, moving faster than ever before, looking increasingly convincing, and infiltrating the very networks where people place their greatest trust. For immigrant communities across the United States, this presents a particularly acute vulnerability, as misleading claims about citizenship requirements, immigration enforcement, and voting rules collide with language barriers and deep-seated fears of government scrutiny. The stakes have never been higher, and the challenge facing journalists and community organizations has fundamentally shifted. No longer is the simple identification of false claims sufficient; the real battle now lies in reaching voters with reliable information before misleading narratives can crystallize into fear-driven decisions. Artificial intelligence has added yet another layer of complexity to this already intricate landscape, making manipulated images and videos easier to produce while simultaneously creating new sources of election information that can themselves generate inaccurate answers. The convergence of these factors has created an environment where the truth often struggles to keep pace with the fiction.

The problem of misinformation itself is hardly new, as Cameron Hickey, CEO of the National Conference on Citizenship, pointed out during a recent media briefing. Rumors, exaggerations, and outright lies have circulated throughout human history, spreading through word of mouth and community gatherings. What has changed dramatically is the speed and scale at which false claims can now travel. A piece of misinformation that might once have taken weeks to spread through a community by word of mouth can now reach millions of people within hours, if not minutes. Social media platforms have supercharged this acceleration, while generative AI technology has made it possible to produce convincing images, videos, and narratives with astonishing speed and minimal cost. Hickey drew important distinctions between various forms of false information: misinformation, which may be spread without malicious intent; disinformation, which is deliberately created to deceive; and conspiracy theories, manipulated context, and misleading framing that all contribute to the same toxic information environment. For voters, however, Hickey emphasized that the intent behind false claims may matter far less than their outcomes. Whether spread intentionally or unintentionally, the results are often identical: people become afraid to vote, afraid to participate in civic life, and make decisions based on false premises that could shape the outcome of elections and the direction of the country.

The economic structure of social media platforms has created perverse incentives that help drive this problem. These platforms are designed to reward engagement above all else, and content that generates anger or fear consistently attracts more attention and interaction than measured, factual information. This creates a dangerous feedback loop where the most emotionally charged content rises to the top, regardless of its accuracy. Furthermore, people are psychologically predisposed to trust information that reinforces their existing beliefs, a phenomenon known as confirmation bias. This combination of factors allows misleading claims to spread rapidly through communities before corrections have any opportunity to catch up. Hickey recommended a practical four-step approach to combating this problem, a method known as SIFT. The first step is to pause before reacting emotionally to any claim. The second involves investigating the source of the information. Third, people should seek out better coverage or independent confirmation from multiple sources. Finally, they should trace claims back to their original source whenever possible. This approach becomes more complicated when artificial intelligence enters the picture, as people increasingly turn to AI systems to verify information. However, Hickey cautioned that chatbots do not necessarily distinguish reliable sources from unreliable ones better than humans can, meaning the same verification principles must apply even when AI is being used as an intermediary for information seeking.

One of the most consequential themes in election misinformation involves claims about federal immigration agents and polling locations. Hickey highlighted how AI-generated images have circulated showing immigration officers intimidating voters, some depicting events that never actually occurred but still influence the people who view them. The existence of real concerns about immigration enforcement makes these fabricated images more believable, creating a chilling effect that can discourage participation even among people who are legally eligible to vote. Misinformation about noncitizen voting has emerged as another significant problem, with claims suggesting that large numbers of noncitizens are casting ballots circulating widely despite their inaccuracy. Citizenship-verification systems themselves can create confusion when records are incomplete or outdated, adding another layer of complexity to an already charged issue. For immigrant voters, a rumor does not need to be technically sophisticated to matter. If it persuades an eligible citizen that voting could expose them or their family to immigration scrutiny, it can effectively suppress participation. The fear generated by these claims can be just as damaging as any actual threat, and the psychological impact can echo through communities long after particular rumors have been debunked.

Jinxia Niu of the Chinese Digital Engagement Project at Chinese for Affirmative Action has tracked hundreds of misleading claims across Chinese-speaking communities, observing how disinformation increasingly moves between platforms and languages. In previous election cycles, WeChat served as a major center of political misinformation, but X has grown increasingly important, particularly for right-wing Chinese-language political content. However, WeChat remains significant because information typically originates on X before being translated and circulated through WeChat groups and YouTube. Much of this content begins in English-language political discourse before being translated into Chinese, but translation is rarely a neutral process. Claims may be altered, intensified, or combined with narratives already familiar to Chinese-speaking audiences. One recurring example involves allegations of Chinese interference in U.S. elections, where Chinese-language influencers reinterpret the lack of mainstream coverage as evidence of a cover-up rather than as an indication that the claims lack merit. Language gaps can then become part of the disinformation itself, with false claims exploiting cultural differences around identification requirements. Niu noted that some Chinese immigrants come from political systems where presenting identification for voting and government services is routine, making U.S. election rules appear confusing or suspicious. This cultural difference creates fertile ground for false claims suggesting that if every voter is not required to present the same form of identification, then anyone can vote, a misleading oversimplification of how electoral systems actually function.

The challenges surrounding artificial intelligence and language gaps are particularly concerning for both Chinese- and Spanish-speaking communities. Niu raised alarms about how users asking election questions in Chinese can receive significantly different source material than those asking the same questions in English. In some cases, AI systems prominently surface Chinese-language outlets that have previously circulated misleading political content, and community members are increasingly using chatbots directly instead of reading traditional news sources. This raises broader accountability questions about which sources AI systems choose to elevate, particularly in languages with smaller pools of high-quality digital journalism. Laura Zommer of Factchequeado has observed similar concerns in Spanish-speaking communities, noting that AI-generated political videos have become considerably more common. Her research testing major AI systems with the same election questions in English and Spanish found that Spanish-language answers were more likely to be incomplete, outdated, or outright incorrect. In one basic example, a chatbot twice gave the wrong date for the U.S. midterm election when asked in Spanish, demonstrating that the problem extends even to fundamental information. Spanish-language responses also tended to cite fewer sources, a significant issue since source quality heavily influences answer reliability. The expansion of AI in campaign advertising compounds these problems, with AI-generated political content now capable of replicating voices, faces, and realistic scenes with little indication of alteration. Disclosure requirements vary by state, and voters may not always recognize artificially produced content, creating particular challenges during casual social media use.

The most effective defense against misinformation ultimately lies with trusted messengers and quality local journalism. When misinformation relies on fear, traditional fact-checking alone may not suffice, and corrections are often more effective when they come through people audiences already know and trust. Community leaders, friends, and family members often serve as the most powerful conduits for accurate information, particularly in closed communication environments such as WhatsApp or WeChat groups where rumors may be shared by relatives rather than strangers. Zommer noted that people may not always change their underlying beliefs after seeing a correction, but they can change what they do with information, and journalists can focus not only on whether a claim is true or false but on explaining how and why it is being used. The speakers repeatedly emphasized the role of local and ethnic media in filling information gaps, particularly as voters increasingly turn to AI systems for election information. Journalists should publish clear, accessible reporting that directly answers common questions, creating sources that AI systems can cite and that voters can find through search engines and community networks. This makes basic service journalism—covering where to vote, when to vote, what identification is required, and what rights voters have—increasingly important. The stakes extend beyond Election Day, as false claims about voting machines, ballot counting, and election fraud continue circulating while results are tabulated. For immigrant communities already navigating language barriers and fears surrounding enforcement, credible information may be as important after polls close as before they open. The central challenge is not simply determining who produces the last word in an online argument; it is ensuring that reliable information reaches voters before fear, mistrust, or manipulation makes the decision for them.