Fact Check-Experts dispute claims that GPS data leads to election fraud by helping cast illegal ballots

Experts told Reuters that while GPS phone data could hypothetically help to identify voters’ approximate movements and their political affiliations, there is no evidence this data is being used to cast illegal ballots. Still, social media users warned voters not to take phones to polling sites during the U.S. midterms by sharing a misleading image that claims GPS data and geolocation services can facilitate casting fraudulent ballots.

One Facebook user sharing the image said, “Do NOT bring your CELLULAR phone into the voting center” (here). More examples of the image, which says GPS data can help “tip the election/s to their preferred candidates” (without identifying who “they” are), are viewable (here), (here).

The image says that GPS data from smartphones and geolocation from cell towers can provide information on whether an individual went to a polling station, and combined with Google user data, could hypothetically determine their voter preferences. It then fuels a narrative that aggregated user data and GPS data can determine who hasn’t voted - “who remains outstanding as a voter” - and that fraudulent votes will be cast on behalf of those outstanding voters. To prevent this from happening, the message urges voters not to bring their cellphones to any polling place.

Experts told Reuters that while it is “certainly believable” that GPS data could predict voter preferences based on their previous online tendencies, claims that this will directly correlate to election fraud by casting fake and illegal ballots stretch the bounds of publicly available evidence.

Todd Humphreys, a professor at the Cockrell School of Engineering at The University of Texas at Austin (here), told Reuters that the claims “have some truth in them, but the final conclusion” – being that geolocation data can be used to create fraudulent ballots – “is bogus.”


While limited in its accuracy, GPS could help to determine whether an individual is inside a polling station, experts told Reuters, especially if the person is using multiple cellphone applications that also try to identify the user’s location.

GPS “provides users with positioning, navigation, and timing (PNT) services” and is operated by the U.S. Space Force ( While GPS satellites “broadcast their signals in space with a certain accuracy,” user accuracy “depends on additional factors, including satellite geometry, signal blockage, atmospheric conditions, and receiver design features/quality” (here).

For GPS-enabled smartphones, accuracy is generally within a 16 feet radius “under open sky” and high-end users can boost accuracy using dual-frequency receivers or augmentation systems. But GPS accuracy is degraded by satellite signal blockage (due to buildings, bridges and trees), indoor and underground use, as well as signals reflected off buildings, among other factors (here).

Mark Psiaki, a professor at the department of aerospace and ocean engineering at Virginia Tech (here), told Reuters that in metropolitan areas with tall buildings, it can be “a lot harder” for precise GPS tracking because “you might see the minimum of four satellites you need,” but the geometry of those satellites is not ideal.

“You might see four (satellites) overhead, and that helps, but none of them are east or west of you because of the buildings in the way,” Psiaki said, which makes determining locations accurately more difficult. He added that GPS might identify a user near a polling station but may not be able to determine whether the voter actually entered the polling site, for example.

However, Aaron Striegel, a professor of computer science and engineering at the University of Notre Dame (here), told Reuters that it might be possible to tell if someone is inside the polling place under certain conditions. An average smartphone uses a combination of methods for mobile location information, including cell site location records (CSLR), GPS and WiFi, to provide a fairly accurate location at a range of 30 feet both indoors and outdoors, he said. That accuracy is enhanced when individuals use social media or games on the phone because many of the ads try to determine the user’s location.

“So, if you are waiting in line to vote, browsing social media on your phone, and the voting area is reasonably big, it is definitely possible to have a reasonable guess that you were voting,” Striegel said. Still, he added, the accuracy would break down in terms of being able to identify if a voter actually visited a ballot box as opposed to being at a polling location, as “that kind of accuracy is not there.”


The National Conference of State Legislatures (NCSL) outlines how states verify ballots, including by signatures, voter records or signed affidavits (here), (here). There is no mention of using online surveillance via GPS and geolocation to verify who voted, contrary to the social media claims.

If there were an effort to cast fraudulent ballots on behalf of outstanding voters, it would “essentially have to be perfect,” Striegel said. Early voting, mail-in or absentee ballots, ballot boxes and vote centers would make this a particularly challenging feat, he added.

“If you mess up in your prediction and try to cast a vote (or) create fraud, that becomes a potential way to be detected,” Striegel said. Some examples are if “person X just votes later in the day or person Y requests an absentee ballot this year.”

“The phone does not necessarily introduce anything new per se in terms of a vulnerability that could not have already been done (see who voted, insert fake votes),” Striegel said. He added that the idea of cellphones facilitating fraudulent ballots has “many trip points due to voting security protocols.”


The circulating image correctly says that Twitter migrated its data processing operations to Google Cloud. The company started working with Google Cloud in 2018 and has since expanded its partnership (here), (here).

The idea that Google is capable of profiling voters as described in the claim is also plausible, according to Psiaki.

“It’s perfectly believable that Google, if they have permission to register my location, they’re certainly collecting all sorts of data,” Psiaki said. For example, by monitoring the sites an individual visits or the posts they create on social media, Google “can have some idea on” voter preferences.

Striegel, too, said that voter preferences, which can come from mining social media and what individuals like or share online, are likely accurate, “but not perfect as all of us have our own quirks and voting thoughts.”

Google did not respond to Reuters requests for comment.

At the Washington Ideas Forum in 2010, Google’s then-CEO Eric Schmidt said that with users’ permission, the company knows “where you are,” “where you’ve been” and is able to “more or less guess what you’re thinking about” (here).

However, Psiaki questioned the logic in the claim: “If you don’t carry your phone to the polling station, then Google might think you didn’t vote when you did.” If an imposter then casts a fraudulent vote and the ballot is accepted, election officials would notice, and “there would be people accused of voting twice if that happened.”

“In that scenario, if you bring your cellphone with you to the polling station, that’s (going to) guarantee that this didn’t happen,” Psiaki said, adding that these claims describe a scenario that is “a bridge way too far.”


Misleading. While aggregated user data might help determine individual voter preferences, there is no evidence of online surveillance including GPS and geolocation services being used to cast fraudulent votes.

This article was produced by the Reuters Fact Check team. Read more about our work to fact-check social media posts here .