DEMOCRAT ELECTION FRAUD BOMBSHELL - EYE-POPPING VIDEO CATCHES EVERYTHING
Milwaukee’s USB Error Delayed a Razor-Thin Primary—But Officials Say the Votes Themselves Never Changed
Election administration has a credibility problem that is harder to solve than either side’s political slogan.
In a close race, even a simple procedural mistake can create hours of uncertainty and give suspicious voters another reason to question the system.
But a mistake in reporting results is not automatically a mistake in counting votes.
That distinction is at the center of the latest controversy in Milwaukee.
Results from Wisconsin’s Aug. 11 primary were delayed after Milwaukee election workers mistakenly exported audit logs instead of election-result files from five of the city’s nine absentee-ballot tabulators.
The mistake came at the worst possible time politically.
The Democratic gubernatorial primary was extremely close, with Milwaukee County Executive David Crowley and state Rep. Francesca Hong separated by only a small number of votes as the final Milwaukee absentee results were still outstanding.
National television networks were waiting for the city’s numbers. Campaigns were waiting. Viewers were waiting.
Then election officials announced that five flash drives brought from Milwaukee’s Central Count facility to the Milwaukee County Courthouse did not contain the result files the county needed to upload.
They contained audit logs instead.
Milwaukee Election Commission Executive Director Paulina Gutiérrez called it “human error.”
“We’re all human, and humans can make errors,” Gutiérrez said during the overnight news conference.
That explanation has not satisfied everyone.
Republican lawmakers and election-integrity critics have argued that an error repeated on five separate tabulators raises legitimate questions about training, supervision and quality control.
Former Milwaukee Election Commissioner Claire Woodall made a similar point in an interview with Votebeat.
“The fact that this happened five times seems more like a training issue,” Woodall said.
“Was this person trained on how to export results? Are they familiar with the machines and how you export results?” she asked.
Those are fair operational questions.
They are also different from the much larger claim now circulating in some political commentary: that the incident shows votes were lost, altered or manipulated.
The evidence released so far does not support that conclusion.
The Milwaukee Election Commission said the ballot count itself had already occurred and the election results remained intact on the tabulators throughout the delay.
“The issue delayed the reporting of Milwaukee’s results, but it did not affect the results,” the commission said in its Aug. 12 explanation.
“No ballots were lost or changed, or needed to be recounted or retabulated.”
That is the most important technical point in the story.
The flash drives were being used to transport already-tabulated result files from Milwaukee’s Central Count operation to the county courthouse, where those files could be opened and uploaded for public reporting.
On five machines, workers selected the audit-log export rather than the results export.
An audit log is not a vote tally. It is a record of activity and events on the tabulator.
So when county staff tried to access the five drives, they did not find the complete files needed to report those machines’ results.
The votes themselves had not disappeared.
They were still stored on the tabulators back at Central Count.
That forced Gutiérrez and other staff members to return to the facility with members of the bipartisan Board of Election Commissioners and a Milwaukee Police Department escort, export the correct files and transport them back to the courthouse.
The commission says the correction produced an approximately one-hour reporting delay. Television coverage stretched the wait longer, with NBC’s Steve Kornacki complaining on air after roughly 90 minutes of uncertainty.
“We’re kind of indefinitely waiting for them to get their act together here,” Kornacki said.
“We waited 90 minutes for them. They did this whole procession, and then we find out there’s nothing on the sticks.”
The phrasing captured the frustration of election night, but “nothing on the sticks” should not be understood literally.
The five drives contained audit logs; what they lacked were the election-result files the county needed at that moment.
There is also some confusion around how many ballots were caught up in the delay.
The Associated Press reported that Milwaukee’s delayed absentee batch involved about 28,000 ballots overall. Fox News, citing Votebeat, reported that roughly 15,000 votes were still awaiting reporting at the point the five affected drives had to be corrected.
Those numbers describe different parts of the same late-night bottleneck, not 15,000 votes that vanished from the election system.
Once the correct result files were retrieved, the commission says it performed another check.
Officials compared a report listing ballot totals by ward from the first process with the final report and said the totals were identical. Media representatives were present for that comparison.
Milwaukee County Clerk George Christenson emphasized that point once the data was finally uploaded.
“At the end of the day, what’s important is that we’re transparent, that the results are secure and that we’re open and honest,” Christenson said.
Crowley was later projected the winner of the Democratic primary over Hong by only a few thousand votes, making the delay politically significant even though officials say it was not outcome-changing.
That narrow margin explains why critics are unlikely to dismiss the mistake as trivial.
Election systems do not operate only on whether the final arithmetic is correct. They also depend on procedures that allow the public to understand why the arithmetic should be trusted.
Five identical export mistakes are therefore worth investigating even if every vote remained intact.
The commission’s own explanation points to an unusual feature of the process.
Material exported from the tabulators is encrypted and, according to the commission, can only be opened by the Milwaukee County Election Commission at the courthouse.
That means city election staff cannot simply plug each drive into another computer before leaving Central Count and open the encrypted file to verify that it contains the correct results.
Republican state Rep. Scott Krug criticized what he called “complete incompetence” and argued the mistake should have been double-checked before officials left the facility.
The commission’s response is effectively that its current security design does not permit that kind of file-content check at Central Count.
That does not end the criticism.
It shifts the question from whether officials should have opened the encrypted files to whether the workflow, training or tabulator interface should contain another safeguard capable of catching the wrong export before transportation begins.
Video from Central Count reportedly showed the tabulator screen displaying that an audit log was being exported, which adds force to the argument that better training or another procedural check could have prevented the mistake.
The commission says it will review its protocols, staffing and practices before the November election.
That review matters because Milwaukee has dealt with other high-profile election-night procedural mistakes.
In 2024, the city restarted the tabulation of more than 30,000 absentee ballots after officials discovered that doors on some tabulators had not been properly secured. The decision delayed reporting but was taken so the affected ballots could be retabulated under corrected procedures.
In 2020, then-election director Claire Woodall-Vogg realized after leaving Central Count that one flash drive containing absentee result data had been left in a tabulator.
A staff member retrieved it and gave it to a police officer, who delivered it shortly afterward. Election officials said timestamps and chain-of-custody information showed the drive had not been altered and that the incident did not affect the validity of the results.
Those episodes make a legitimate argument for stronger process discipline in Milwaukee.
They do not, by themselves, establish election fraud.
That is also where the connection to the 2020 presidential election needs to be handled carefully.
After that election, a joint statement from federal, state and election-industry officials described Nov. 3, 2020, as “the most secure in American history.”
That sentence has often been treated in political debate as though officials were claiming no election worker anywhere made a mistake.
The statement itself was narrower.
It said there was no evidence that any voting system deleted or lost votes, changed votes or was in any way compromised, while also pointing to paper records and other safeguards used to identify and correct errors.
The Milwaukee incident this week does not retroactively prove or disprove that assessment.
It occurred in a different election six years later, involved different personnel and concerned the export and transportation of already-tabulated result files.
Nothing released by Milwaukee officials, the county or independent news organizations has shown that the Aug. 11 error changed a vote.
That does not mean voters should simply ignore it.
The better lesson is narrower and more useful.
Milwaukee’s safeguards appear to have caught the problem: county staff could not upload the wrong files, officials returned under police escort, the correct results were exported and the ballot-total reports were compared before the results were posted.
But safeguards that catch an error after a late-night trip across the city are not a substitute for procedures designed to prevent five copies of the same error in the first place.
The distinction matters because election confidence is damaged in two different ways.
Actual manipulation would undermine the result itself.
Repeated avoidable mistakes can undermine public confidence even when the result remains accurate.
Milwaukee officials have evidence for the second problem. They have not produced evidence of the first, and neither have their critics.
The Aug. 11 episode therefore supports a strong case for better training, clearer export procedures and public scrutiny before November.
It does not support a factual claim that 15,000 or 28,000 votes went missing, that the machines changed ballots or that the Crowley-Hong result was manufactured.
The primary ultimately settled who Democrats will send into the governor’s race against Republican Tom Tiffany.
What the night did not settle is whether Milwaukee can redesign its process well enough that the next close election is remembered for the voters’ decision rather than for another preventable procedural error.
For election officials, that may be the more important challenge.
May you like
Trust does not require pretending mistakes never happen.
It requires showing exactly what went wrong, proving what did not change and fixing the process before the mistake happens again.
