Two warnings, one message: verify before you click

With the 2026 United States midterm elections now weeks away, American authorities have issued a pair of warnings that share the same core message: scammers and hackers are using artificial intelligence to make election fraud harder to spot, and voters and election officials alike should slow down and verify before they click, donate, or share personal information. The warnings, trending per news coverage on Friday, come from the FBI's Internet Crime Complaint Center (IC3) and from a joint alert by the FBI and the Cybersecurity and Infrastructure Security Agency (CISA).

The timing is deliberate. Election season is when public attention, political passion, and money all peak at once. Campaigns send a flood of legitimate fundraising texts and emails, volunteers knock on doors, and voters search for registration deadlines and polling places. That noise is exactly what fraudsters hide in. The FBI says these schemes have targeted voters in previous state and local elections, and that AI tools have made the impersonations far more convincing than they were even two years ago.

This article explains what the two alerts say, the specific schemes officials describe, why AI changes the threat, and the practical steps voters and campaign workers can take to protect themselves.

What the FBI's warning says

The FBI's public service announcement, published through IC3, warns that scammers could exploit the 2026 US midterm election to carry out financial fraud schemes aimed at victims across the country. It updates a similar warning issued ahead of the 2024 general election. The central point is that criminals are borrowing the identities of real candidates: their names, images, logos, and slogans. With those borrowed identities, they solicit campaign contributions that never reach a campaign, sell merchandise that is never delivered, or harvest personally identifiable information that can later be used for identity theft and further fraud.

The FBI is explicit that AI is the force multiplier. Scammers use AI tools to generate highly realistic voices, images, and written messages that impersonate candidates, campaigns, government entities, and even people the victim personally trusts. The realism, the bureau says, makes the schemes harder to detect and increases the chance of financial loss or exposure of personal data. A phone call that sounds like a candidate's voice, a text that reads like a campaign volunteer's, or a social media ad carrying a familiar logo can no longer be taken at face value.

Officials have not tied the warning to a single confirmed wave of cases in this election cycle. The alert is preventive: it describes schemes that have been seen before and warns that the same playbook, sharpened by AI, is likely to be reused as voting approaches. That distinction matters. The warning is not a claim that any particular campaign has been breached or that any specific election result is in doubt. It is a fraud alert aimed at protecting individuals.

What the joint FBI–CISA alert adds

The second alert, issued jointly by CISA and the FBI and reported by Reuters on Friday, is aimed more directly at the people who run elections. It urges election officials, candidates, and their political organizations to be on guard against AI-powered tricks ahead of the midterms. According to the reporting, the alert says hackers may try to disrupt the midterms using bogus text messages and emails designed to steal passwords or other sensitive data.

The alert describes election officials, candidates, and their political organizations as attractive targets for malicious cyber criminals. It also says that increasingly capable AI — described in the alert with the administration's term "super intelligence" — is allowing attackers to sharpen their rogue emails and texts so they look and sound more convincing. In practice, that means phishing messages that once carried telltale grammar errors or generic greetings can now be tailored to a specific office, a specific staffer, or a specific local procedure.

Together, the two alerts cover both ends of the election ecosystem: the public-facing fraud that targets voters' wallets and data, and the back-office phishing that targets the accounts and systems campaigns and election offices rely on.

The four schemes the FBI describes

The FBI's announcement lays out four recurring schemes in detail. Understanding each one makes the red flags easier to recognize.

1. The campaign investment pool scheme. Scammers promise victims a guaranteed return on campaign contribution dollars by investing in a supposed candidate campaign pool that pays out after the candidate wins. Victims are also encouraged to recruit others to invest, increasing their own share of the payout. The FBI's response is blunt: campaign donations are not an investment. They will not increase in value, and they will not be returned. Any pitch that treats a donation as an investment with a guaranteed payout is a fraud.

2. The fake PAC scheme, also known as "scam PACs." Scammers persuade victims that they are donating to a legitimate Political Action Committee, such as a Super PAC or a candidate's campaign committee, but the scammers have no affiliation with that committee and keep the funds. The FBI notes that some scam PACs may even appear registered, and advises checking how a committee spends its money: a PAC that reports high expenses for fundraising, consulting, compliance, or other administrative costs may be a scam.

3. The online purchase scheme. Scammers claim to sell political merchandise — shirts, hats, yard signs, and similar items — but never ship the purchase. This is a classic non-delivery fraud, dressed in campaign branding. Before buying, the FBI advises researching the seller through reviews and complaints, for example via the Better Business Bureau.

4. The voter registration scheme. Victims receive a text or email stating they are not registered to vote in their state, with a link to a supposedly official state voter registration page. The page is fake, and its purpose is to steal personal information for identity theft and to set the victim up for further fraud. The FBI advises checking registration only through the official Vote.gov portal rather than through links in unsolicited messages.

Why AI makes this election cycle different

Election scams are not new. What has changed is the cost and quality of impersonation. Cloning a voice once required studio equipment and skill; today, AI tools can produce a convincing voice from a short sample of recorded speech. Writing a persuasive fundraising email once took effort and left stylistic clues; AI writing tools can now produce fluent, personalized messages at scale. Generating a realistic image or a fake campaign logo once took a designer; now it takes seconds.

That shift lowers the barrier for criminals and raises the burden on the public. The old advice — "look for typos" or "listen for a robotic voice" — is no longer enough on its own. A message can be grammatically perfect and still be fake. A voice can sound warm and familiar and still be synthetic. The FBI and CISA alerts both point, implicitly, to the same conclusion: verification must move from judging how a message looks or sounds to checking it through an independent channel.

The broader threat landscape reinforces the point. Election-related threats this cycle have not been limited to online fraud: authorities have separately investigated more than 100 suspicious letters sent to US election offices and senators in the weeks before the midterms. Campaigns are also in their final, high-intensity stretch — with national figures touring battleground states — which means voters are receiving more political messages than usual, legitimate and otherwise. And the same AI-driven fraud techniques appearing in election scams are surfacing in other sectors too, as seen in recent cyberattacks on financial institutions that exploited trusted digital channels.

How to protect yourself

The FBI's advice is practical and applies well beyond election season:

  • Treat unsolicited contact with caution. Be wary of unexpected calls, texts, emails, or surveys. Do not provide personal information to people you do not know, and do not click unknown links.
  • Donate through official channels only. If you want to support a candidate or committee, go to the campaign's official website directly — type the address yourself or use a known, trusted link — rather than responding to a text, email, or social media ad. Verify the committee's registration and financial filings on the Federal Election Commission website.
  • Remember that donations are not investments. No legitimate campaign offers a guaranteed return, a payout after victory, or a bonus for recruiting other donors. Those promises are hallmarks of fraud.
  • Check registration at Vote.gov. If a message claims you are not registered, ignore its link and check your status through Vote.gov or your state election office's official site.
  • Verify before you act. If a message claims to be from a candidate, campaign, or election office, confirm it through a separately obtained phone number or official website. For campaign staff, confirm unusual requests — especially anything involving passwords, payments, or sensitive files — through a second channel.
  • Use strong account protections. Election officials and campaign workers should use multi-factor authentication on email and campaign accounts, be cautious with attachments and links, and report suspicious messages to their IT or security contacts promptly.

What to do if you have been targeted

The FBI asks anyone who encounters an election-related scam to report it to their local FBI field office or to IC3 at ic3.gov, with as much detail as possible: who contacted you, the phone number, email address, or website used, how payment was requested or made, what you were told, and what you provided. Reports help investigators spot patterns, warn others, and, in some cases, recover funds or shut down fraudulent operations. Even if no money changed hands, reporting a suspicious message can help authorities track an active scheme.

If you shared personal information, consider the standard identity-theft precautions: monitor financial statements, consider a credit freeze or fraud alert, and change passwords on any accounts that may have been exposed. If you clicked a suspicious link, run a security scan on your device and change the passwords of important accounts from a clean device.

Why midterms attract this kind of fraud

Midterm elections draw intense national attention while remaining highly decentralized. Thousands of federal, state, and local races run simultaneously, each with its own campaign, its own fundraising operation, and its own communications. That fragmentation gives scammers an enormous surface to imitate: a voter may have no way of knowing what a real message from a particular local campaign looks like, because they have never received one before.

Turnout dynamics add to the risk. Voters who participate only in presidential years may be less familiar with midterm procedures, deadlines, and registration checks — exactly the uncertainty that the fake voter-registration scheme exploits. And the emotional temperature of campaigns makes urgency a powerful weapon: a message claiming a donation deadline expires tonight, or that a voter's registration is about to be cancelled, pushes people to act before they verify.

The official alerts are a reminder that the most effective defense is unglamorous: pause, verify through an independent channel, and never let urgency override caution. The technology of fraud has changed, but the antidote — checking the source before trusting the message — has not.

Sources: FBI Internet Crime Complaint Center (IC3) public service announcement, 15 Oct 2026; Reuters reporting on the joint FBI-CISA alert, 16 Oct 2026; Help Net Security; Money.

The interference fears are not only digital: the US Embassy in Ottawa has accused Canada of trying to influence the 2026 midterms, escalating the warnings into a diplomatic row.