As the November 3 midterms approach, misleading messages about immigration, voting rights, and AI-generated political content are raising fears in immigrant communities. Experts warn that even eligible citizens could be discouraged from voting.
Magazine, The Immigrant Experience
Imagine becoming an American citizen, finally earning the right to vote, and then receiving a message that makes you afraid to go to the polls.
Perhaps it comes through a family WhatsApp group. Someone warns that immigration officers will be waiting at voting locations. Another shares a video claiming voting machines are changing ballots. A relative forwards a message suggesting that naturalized citizens could face trouble if they vote.
You trust the person who sent it. They may even believe they are protecting you.
But what if the information is false?
With the November 3, 2026, midterm elections approaching, this is the kind of fear election misinformation can create. And for immigrant communities, where information often travels through family, friends, cultural networks, and social media, the consequences can reach far beyond a misleading post.
A citizen who has every right to vote may decide to stay home.
That possibility was at the heart of an American Community Media briefing, Who Gets the Last Word? Election Disinformation in Immigrant Communities, moderated by associate editor Pilar Marrero.
Three experts—Cameron Hickey, CEO of the National Conference on Citizenship; Jinxia Niu of Chinese for Affirmative Action; and Laura Zommer, co-founder and CEO of Factchequeado—examined how misinformation spreads, why immigrants can be particularly vulnerable, and how artificial intelligence is making it harder to know what to believe.
Their findings point to a growing concern: The danger is not simply that people may believe something untrue. It is that fear and confusion could keep eligible voters from participating in an election that affects their families and communities.
When Fear Keeps Eligible Voters Away
Cameron Hickey, who also founded the National Conference on Citizenship’s Algorithmic Transparency Institute, began by reminding journalists that misinformation is nothing new.
Rumors, misleading claims, and deliberate lies existed long before social media. What has changed is how quickly they spread and how easily technology can make them appear believable.
Today, a false claim can travel across the country in minutes. A convincing video can be created without the event ever taking place. And a message designed to frighten people can reach thousands before anyone has checked whether it is true.
“This isn’t just a media problem,” Hickey told journalists. “This is a civic health problem.”
He described AI-generated images showing Immigration and Customs Enforcement (ICE) agents intimidating voters near polling locations. According to Hickey, researchers have documented examples of fabricated imagery circulating online.
He also cautioned that real concerns about immigration enforcement must be distinguished from manufactured images. Both can contribute to anxiety, but journalists have a responsibility to establish what actually happened.
For immigrant families already concerned about immigration policies and enforcement, a frightening image does not necessarily need to be real to have an impact.
Someone who recently became a citizen may wonder whether going to vote could expose them or their family to unwanted attention.
That is what makes these messages so harmful.
Hickey also raised concerns about allegations involving noncitizen voting and the federal government’s use of the SAVE citizenship-verification system. He warned that incomplete or outdated records could cause newly naturalized citizens to be incorrectly identified as noncitizens.
The federal government uses citizenship information from multiple agencies, but changes in immigration status may not always appear immediately across every database.
For a newly naturalized citizen, being mistakenly questioned about eligibility can be unsettling. When such claims are amplified by public officials or widely followed personalities, they can create even greater uncertainty.
Hickey emphasized that journalists must examine these claims carefully rather than treating their circulation as proof.
For voters, the message is equally important: A rumor about immigration enforcement or voting eligibility is not a substitute for the law or official election information.
How False Information Travels Through Immigrant Communities
For many immigrant families, staying connected to relatives and communities means relying on familiar communication networks.
WhatsApp groups connect families across borders. WeChat brings Chinese-speaking communities together. YouTube, Facebook, TikTok, and X help people follow events in their countries of origin and in the United States.
These connections are valuable. They help communities share news, maintain relationships, and support one another.
But they can also carry misinformation.
Jinxia Niu, who leads the Chinese Digital Engagement Program at Chinese for Affirmative Action, explained how misleading political narratives frequently begin in English before being translated and shared within immigrant communities.
Her organization’s research has tracked more than 600 pieces of misinformation and disinformation.
Niu said that while WeChat remains an important platform, X has become a major source of Chinese-language election narratives, some of which are later circulated through WeChat groups and YouTube.
By the time a message reaches a family or community network, it may have been translated, rewritten, or mixed with additional claims.
And because the message arrives in a familiar language, sometimes from someone the recipient knows personally, it may be accepted without much questioning.
Niu pointed to a recurring false narrative about noncitizen voting.
In China, identification is required for many everyday activities. For immigrants accustomed to that system, the fact that the United States does not have one national voter identification requirement can seem confusing.
Disinformation campaigns exploit that difference by suggesting that if voters are not universally required to show identification at polling places, anyone can vote.
That is not how American elections work.
Federal law prohibits noncitizens from voting in federal elections, and states have voter registration and election administration systems to determine eligibility. Identification requirements vary depending on state law.
But simply repeating those facts may not be enough.
For immigrants who grew up under different political and electoral systems, understanding American elections can require explanations that recognize their experiences.
Niu described how claims about illegal voting, when combined with immigration enforcement fears, can create anxiety even among naturalized American citizens.
Her organization’s research has also identified anti-immigrant narratives affecting Chinese, Indian, Filipino, and Vietnamese communities.
These findings are central to its forthcoming report, Divided by Design, which examines how fear-driven political misinformation can deepen divisions within Asian American communities.
The challenge is not limited to one language or community.
A message that works by exploiting cultural differences in one immigrant population may take a different form in another.
That is why meaningful fact-checking cannot depend on translation alone. It must also understand the community receiving the information.
When AI Gives Different Answers in Different Languages
Perhaps one of the most troubling revelations during the briefing concerned the growing number of people using artificial intelligence to answer questions about voting.
More Americans are asking AI chatbots about candidates, election dates, registration requirements, and voting procedures.
For people who prefer to communicate in languages other than English, these tools can appear especially helpful.
But what happens when the answers are different depending on the language used?
Laura Zommer, co-founder and CEO of Factchequeado, a nonprofit working to combat Spanish-language misinformation in the United States, shared findings that deserve serious attention.
She cited research by the Institute for Strategic Dialogue, which tested six major AI chatbots using election-related questions in English and Spanish.
The study found that 29% of responses to basic election questions in English were inaccurate, incomplete, or outdated.
When those questions were asked in Spanish, the rate rose to approximately 45%.
Think about what that means for someone trying to understand how to participate in an election.
A person asking about voting in English could receive one answer, while someone asking the same question in Spanish might receive information that is incomplete or incorrect.
And these were not always complicated legal questions.
In one example, Meta’s AI chatbot twice gave the wrong date for the 2026 midterm elections, answering November 4 instead of November 3.
Imagine someone relying on that answer and showing up to vote a day late.
The research also found that AI chatbots provided fewer supporting sources in Spanish than in English, making it harder for users to check where the information originated.
Niu raised similar concerns about Chinese-language AI responses, warning that the sources surfaced by chatbots may differ significantly depending on the language used.
For immigrant communities, this creates another barrier to reliable information.
Technology that promises to make knowledge more accessible should not leave people with less accurate answers because of the language they speak.
And for journalists serving multilingual audiences, it raises an important responsibility: Reliable election information must be available in the languages communities actually use.
AI Political Videos Are Becoming Harder to Recognize
Zommer also warned that AI-generated political content is becoming increasingly sophisticated.
During the 2024 elections, researchers documented AI-generated images circulating online. In 2026, she said, videos created or enhanced with artificial intelligence have become a much more prominent part of political messaging.
Some feature realistic portrayals of candidates or events that never happened. Others use AI to alter existing footage or create persuasive political advertisements.
Citing research from the Wesleyan Media Project, Zommer said researchers had identified nearly 170 political advertisements using artificial intelligence, representing approximately $80 million in advertising spending.
Even more concerning, 69% of those advertisements did not disclose that AI had been used.
This does not mean every AI-generated advertisement is automatically false or deceptive. But when political content is artificially created or altered without clear disclosure, people may have difficulty knowing what they are watching.
For someone scrolling quickly through social media, distinguishing a real recording from a fabricated one may not be easy.
And as political videos move from one platform to another, their original context can disappear.
A video may be reposted with a new headline, translated into another language, or shared with a warning that encourages viewers to react immediately.
By the time it reaches a family group chat, its origin may be almost impossible to recognize.
Why Trust Matters More Than Ever
During the briefing, journalist Carlos Roa raised a question that gets to the heart of this challenge: If fear is one of the main reasons misinformation spreads, how can journalists communicate facts in ways that people will actually believe?
Niu’s response focused on trusted messengers.
She explained that correcting misinformation often begins with people who already have meaningful relationships within their communities.
That may be a community leader, a local journalist, a friend, or a family member.
Her point was simple: Fear is a natural human response. When people feel threatened, they often turn to those they trust for guidance.
That is why simply publishing a correction may not be enough.
People need information from sources they recognize, in language they understand, delivered with respect for their experiences.
Zommer described a similar approach at Factchequeado, which works with a network of 151 media and community organizations across the country.
Alongside its fact-checking work, the organization listens to conversations taking place in communities, including rumors circulating through WhatsApp groups and everyday interactions.
Those conversations are not always visible to researchers studying public social media platforms.
Understanding them requires something technology cannot replace: Listening to people.
For ethnic and community media, this is an important reminder.
Our responsibility is not simply to report what has been said. It is to help communities understand what is true, what remains unverified, and what they can do with that information.
Before You Believe It, Before You Share It
The experts offered practical advice that every family can use.
Hickey introduced a simple fact-checking method called SIFT, developed by information literacy researcher Mike Caulfield.
It begins with something many of us forget to do when we receive a frightening or emotional message: Stop.
Before forwarding that video, sharing that warning, or accepting that alarming headline, take a moment to ask where the information came from.
Then investigate the source.
Is it an official election office? A credible news organization? Or an account you have never heard of?
Next, look for better coverage.
If an important event has happened, reliable news organizations or official agencies may have information that can help confirm what is true.
Finally, trace the claim to its original source.
A message may have been shared hundreds of times, but that does not make it accurate.
And remember, AI chatbots can make mistakes too. Do not rely on them alone for election dates, registration deadlines, polling locations, or voting requirements.
Official election offices remain the best source for that information.
What You Need to Know Before November 3
The 2026 midterm elections will take place on Tuesday, November 3.
If you are a naturalized American citizen who meets your state’s voting requirements, you have the same voting rights as other eligible U.S. citizens.
Noncitizens cannot vote in federal elections. Some jurisdictions have separate rules for certain local elections, but those exceptions do not apply to federal contests.
If you are unsure about your registration, polling location, identification requirements, or voting options, check with your state or local election office.
You can also use these resources:
- Vote.gov — Find voter registration information and links to your state’s official election resources.
- USA.gov Voting and Elections — Understand voting procedures, election rules, and where to find assistance.
- Factchequeado — Find Spanish-language fact-checks and resources to help identify misleading information.
- California Secretary of State — Find official voting information for California, including voter registration and voting options.
For California voters, the regular online registration deadline was October 19, but eligible citizens who missed it may still use same-day voter registration at designated locations through Election Day.
Most importantly, do not allow an alarming social media message to make the decision for you.
If you encounter misinformation or believe someone is attempting to intimidate voters, seek guidance from your election office or an established voting-rights organization.
And if someone in your family sends information that turns out to be false, correct it with care.
Many people share these messages because they are afraid or because they believe they are protecting someone they love.
A conversation grounded in facts may accomplish more than an argument.
Your Vote Should Never Be Lost to Fear
The November elections will determine who represents communities in Congress and in many state and local offices. Decisions made by those elected officials can affect education, healthcare, housing, immigration policy, economic opportunities, and the everyday lives of families across America.
For naturalized citizens, the right to vote is also an opportunity to participate directly in decisions about the country they now call home.
That right should not be diminished by a fabricated video, a misleading chatbot answer, or an unverified message passed along by someone they trust.
The next time a frightening election message arrives in your family WhatsApp group, don’t rush to forward it.
Ask where it came from. Check the facts. And when you know what is true, share that information with others.
Because sometimes, protecting someone’s right to vote begins with one simple conversation.
Before you believe it, verify it. Before you share it, check it. And if you are eligible to vote, don’t let misinformation make that decision for you.
#Election2026 #ImmigrantCommunities #AIandDemocracy #MediaLiteracy #VotingRights
