Encyclopedia of contested claims
Available evidence from U.S. elections generally indicates that documented fraud involving mail-in ballots is rare, and not meaningfully more common than fraud associated with in-person voting at a scale that would affect typical election outcomes. Mail voting does have different security risks than in-person voting, but states use verification, tracking, and chain-of-custody safeguards to reduce those risks.
9 reviewing models concluded the claim is not supported by the available evidence.
The Adjudged panel has not yet completed its independent review of this claim. This draft summarizes the main lines of evidence, identifies testable sub-claims, and lists source candidates for panel evaluation.
Available evidence from U.S. elections generally indicates that documented fraud involving mail-in ballots is rare, and not meaningfully more common than fraud associated with in-person voting at a scale that would affect typical election outcomes. Mail voting does have different security risks than in-person voting, but states use verification, tracking, and chain-of-custody safeguards to reduce those risks.
The claim asks whether mail-in ballots are more susceptible to fraud than in-person voting. In U.S. elections, “mail-in ballots” may include absentee ballots requested by voters, ballots automatically mailed to registered voters in some jurisdictions, and ballots returned by mail, drop box, or election office.
Fraud can refer to several different acts: voting on behalf of someone else, forging a signature, submitting multiple ballots, coercing a voter, intercepting ballots, or manipulating returned ballots before counting. In-person voting can also involve impersonation, double voting, voting by ineligible individuals, or poll-worker misconduct.
The key question is not whether either method has zero risk. It is whether mail voting has shown higher real-world rates of fraudulent ballots than in-person voting, especially at levels large enough to affect election outcomes.
Research and official post-election reviews generally find that documented voter fraud in U.S. elections is uncommon across voting methods. Cases involving mail ballots do occur, and some election experts note that mail voting creates distinct vulnerabilities because the ballot leaves the controlled polling-place environment.
However, the available record does not support a broad conclusion that mail-in ballots produce widespread fraud compared with in-person voting. States use signature verification, barcode tracking, ballot reconciliation, voter-roll checks, secrecy envelopes, cure processes, and criminal penalties to detect or deter improper ballots.
In-person voting has its own safeguards, such as check-in procedures, poll books, poll observers, voter identification rules in some states, and provisional ballots. It also has its own possible vulnerabilities, including administrative errors, double-voting attempts, and rare impersonation cases.
The most careful comparison is between documented rates, not theoretical possibilities. The documented rate of fraud for both mail and in-person voting appears very low relative to the number of ballots cast, and available evidence does not indicate that mail voting is broadly more susceptible in a way that commonly changes results.
Fraud is difficult to measure perfectly because successful fraud may go undetected. Differences in state rules, enforcement capacity, ballot verification standards, and reporting practices can also make direct comparisons between mail and in-person voting challenging.
Some specific forms of misconduct, such as coercion in a household or improper ballot collection, are more plausible in mail voting than at a polling place. By contrast, some issues often raised about in-person voting, such as impersonation, are also hard to measure but appear rarely documented.
A stronger assessment would require more standardized national reporting on investigated and confirmed cases by voting method, including the number of ballots affected and whether the conduct was intentional, administrative error, or voter misunderstanding.
The umbrella claim is actually several claims bundled into one. Each needs its own evaluation.
| Model | Part 1 | Part 2 | Part 3 | Overall |
|---|---|---|---|---|
| Grok 4.3 | No · 87% | No · 82% | No · 78% | No · 90% |
| OpenAI GPT-5.4 | No · 87% | No · 82% | No · 78% | No · 90% |
| Mistral Medium 3.5 | No · 87% | No · 82% | No · 78% | No · 90% |
| Claude Opus 4.7 | No · 87% | No · 82% | No · 78% | No · 90% |
| Llama 4 Maverick | No · 87% | No · 82% | No · 78% | No · 90% |
| Gemini 3.1 Pro | No · 87% | No · 82% | No · 78% | No · 90% |
| GLM 5.1 | No · 87% | No · 82% | No · 78% | No · 90% |
| DeepSeek V4 Pro | No · 87% | No · 82% | No · 78% | No · 90% |
| Qwen 3.7 Max | No · 87% | No · 82% | No · 78% | No · 90% |
| Kimi K2.6 | — | — | — | Incomplete |
The current evidence leans one way. But we're not committed to the conclusion, we're committed to the evidence.
Evidence from U.S. elections shows documented mail-in ballot fraud remains rare overall and does not occur at rates meaningfully higher than in-person voting when measured against total ballots cast or outcomes...
Mostly false. The best available U.S. evidence does not support the claim that mail-in ballots are generally more susceptible to fraud than in-person voting in a way that is widespread, common, or outcome-deter...
The claim that mail-in ballots are more susceptible to fraud than in-person voting is not strongly supported by available evidence. Documented cases of voter fraud involving mail-in ballots in recent U.S. elect...
Mail-in ballots are somewhat more susceptible to certain types of fraud than in-person voting, but the absolute rate of documented fraud in both methods is very low, and the difference is not large enough to routinely affect election outcomes in U.S. federal or statewide elections. Confidence: Moderate-to-high on the overall framing; high on the claim that documented fraud is rare in both; moderate on the comparative susceptibility question because it depends on how "susceptibility" is defined (theoretical vulnerab...
Mail-in ballots are not meaningfully more susceptible to fraud than in-person voting at a scale that would affect typical election outcomes. Confidence: High The available evidence from various credible sources...
Available evidence does not support the claim that mail-in ballots are meaningfully more susceptible to fraud than in-person voting in a way that affects election outcomes, though they do present different logi...
No, available evidence does not show that mail-in ballots are broadly more susceptible to fraud than in-person voting in a way that meaningfully impacts U.S. election outcomes. While mail-in voting does present...
No, mail-in ballots are not broadly more susceptible to fraud than in-person voting in U.S. elections. While mail voting introduces different security risks—such as ballots being handled outside controlled poll...
No, available empirical evidence does not support the claim that mail-in ballots are broadly more susceptible to fraud than in-person voting at a scale that affects U.S. election outcomes, although the two meth...
