There's nothing quite like a bill that seeks to solve a problem that barely exists. Voter fraud is very rare in America, despite the president's claims, which have a chilly relationship with the truth.
The Save America Act is a tool for voter suppression, not election integrity.
What a fabulous piece of work Gabe. Hugely informative, but hugely depressing that they are so coddled into conformity and laziness. While not mentioned in this piece, I surely agree with docking their luxurious pay/privileges until they fund the folks currently working without pay, like the TSA.
Reading the comments from the senators to Gabe was so disheartening. They are so cut off from "normal" America and don't seem to care about how Americans are affected by their actions. But I don't know if there's any way back from this. Anyone who tries to make a difference quickly discovers how hard it is to swim upstream.
Great work Gabe ! If the Democrats are not being allowed to submit amendments, how can the Senator interviewed really believe they are engaged in a meaningful process. Sounds like: here's the bill like it or not...or I suppose debate could convince Republican Senators to amend the bill, assuming they weren't so scared of the _____________.
This, depressingly, reminded me of what Romney said about being told when he became a senator that only 20 of them do all the work and the other 80 are just along for the ride. 🤦🏻♀️
I keep a copy of the Constitution on my desk. Still, until the Save Act, I thought the requirement that only citizens vote in federal elections was in the Constitution. According to AI (my new best friend), since 1924, the requirement is in either the state constitutions or the statutes of all 50 states. (I know the significance of the 1924 date.) This means to me that states can change that requirement. We also do not have any national registry of citizens. The Save Act to me is an insurance policy. The cost (people will have to locate proof of citizenship) benefit (there is presently little voter fraud) analysis does not resonate with me. A belief in our election integrity is an intangible benefit. I want to confine the benefit to citizens whose first loyalty is to this country and who stand willing to defend it when necessary, not just those who seek the economic benefits.
I hope the conservative contingent does pull off one of those crazy scenarios. Sleeping on the floor and/or standing up and speaking (and not going to the head of the line at airports) would be a good thing.
Rosemary, I genuinely used to enjoy your comments. You are clearly different from me. Different part of the country, different politics, and different career (I seem to recall former lawyer?). Although I appreciate your transparency around using AI to contribute to your comments and thinking, I find myself becoming less trustworthy of your comments and thinking.
I use AI every day in my work, and I want to urge you to tread carefully when using it. I’ve already seen other folks push back on your AI informed arguments. It isn’t the tool we think it is!
Sincerely, from someone who wants to hear your POV, not an AI informed or enforced one.
Thank you. I am still in testing phase with Gemini Google. It is a Godsend for a quick jolt to a flawed memory. And, I am pleased when there is pushback. For eg, I will do a little more research into the source of citizen restriction re: voting. (I think Mr. Miller reads the 14th Amendment too broadly.) It appears that the restrictions on non citizens came in conjunction with immigration policy (and are, coincidentally loosening for the same reason.) I belong to a group--braverangels.org--which fosters civil discourse about politics, etc.
The 14th Amendment, Section 1, is very clear that citizens who are born in the U.S. or who are naturalized are the only people eligible to vote. No state can change that requirement. And since even republicans concede that voter fraud is almost nonexistent, the only purpose of the SAVE Act is to make it hard for people (mostly women, who tend to vote democratic) to prove their citizenship by having to go to the time, effort and sometimes expense to dig up their birth certificates and wedding records to explain why their last name on their drivers license is now different from what’s on their birth records. It’s called voter suppression and it’s as clever as it is evil. Why do you support it?
I hate it when people argue that "women" or "minorities" will be adversely affected as if the class of women or minorities are too addled to obtain documents to prove citizenship. I reject that premise (as a woman) and I would be equally offended by the assertion if I were a "minority." Are you a White Male?
It would be fun to see if they were willing to whittle the SAVE AMERICA ACT down to the basics, on having proper ID for voting and nothing else. That might be popular enough to actually pass. The polls show 81% for it, with a call in from their constituents, it might pass. Heavy on the might.
But why bother spending ANY time on legislation designed to solve a problem that doesn’t exist? Oh, right, so republicans can suppress the vote while appearing to be concerned about “election security.”
I appreciate your view, but the same was said of welfare fraud, and that seems to be proving less than an accurate statement of fact. In Southern California in a recent statewide referendum, the vote tally on the machines was roughly 511,000+, the vote tally that was recorded was 567,000+. Something is wrong there, and have no evidence of things either way, just that there is a stated discrepancy, which means it should be investigated.
As to voting, the Constitution states it is limited to American citizens, shouldn't they have to prove they are such before voting? I am all for having the state or federal government pay for any effort to find birth certificates or whatever else is needed. The question remains, shouldn't there be some proof of citizenship, before one votes?
I decided to check out your first comment using Perplexity. The result is so long I’m not gonna inquire about your second point! It’s interesting though. Here goes:
The claim you quote about a 511,000 vs 567,000 machine‑count discrepancy in a Southern California referendum is almost certainly based on a misunderstanding or misreporting of the underlying data, and as stated it is not strong evidence of large‑scale fraud.[1][2][3]
## On the specific “511,000 vs 567,000” claim
Several recent California controversies look very similar to what you’re describing: citizen groups or activists have alleged large gaps between ballots “logged” or “machine tallied” and ballots “reported,” which local election officials then showed were based on reading preliminary or partial numbers incorrectly.[2][3][1]
In Riverside County, for example, Sheriff Chad Bianco has claimed a roughly 45,000‑vote discrepancy in a redistricting referendum, prompting him to seize hundreds of thousands of ballots for his own hand count. The county registrar, however, explained that when the actual certified records are compared, the difference between ballots cast and ballots counted is about 103 votes, roughly 0.016%—a level consistent with normal administrative variance (spoiled ballots, provisional ballots rejected, etc.). In other words, the headline “tens of thousands of missing votes” arose from misinterpreting raw logs and preliminary machine data, not a real gap in the certified totals.[3][4][1][2]
Your 511,000 vs 567,000 figures fit that same pattern: a large, round discrepancy claim circulating without a clear link to the official canvass reports, likely mixing apples (e.g., one county’s machine totals at a point in time) with oranges (multi‑county or statewide certified numbers, including late‑counted mail and provisional ballots). Without a specific jurisdiction, contest name, and official canvass, the claim by itself is weak evidence—what we can say from similar, well‑documented disputes is that when election officials and independent reviews dig into the details, the “huge gap” tends to collapse to a tiny one explained by normal election processes.[5][1][2][3]
## Does a discrepancy mean “fraud” or “shouldn’t be investigated”?
You’re right on one core point: if there is a genuine, documented discrepancy between ballots cast and ballots counted, it should be investigated. Election systems expect reconciliation: every batch and machine has logs, and canvass procedures are designed precisely to catch and correct mismatches. A healthy system:[5]
- Tracks differences between machine counts, hand counts, and reported totals.
- Explains them in public reports (spoiled ballots, duplicates, provisional ballots rejected, etc.).
- Escalates to audits or recounts when thresholds are exceeded.
That’s different, though, from treating any large number someone quotes online as proof of fraud. The Riverside case shows how quickly a claim of a 45,000‑vote gap can circulate, only for the documented difference to be on the order of a hundred votes once the actual records are examined. So “there is a stated discrepancy” is not the same thing as “there is a verified discrepancy in the official canvass that survived audit.”[1][2][3]
A reasonable standard is:
- Yes, investigate specific, documented mismatches in official records.
- No, don’t presume fraud based on unresolved or third‑hand numerical claims, especially when similar claims have repeatedly been debunked after basic reconciliation work.[6][1][5]
## Welfare fraud as an analogy
On welfare and government‑benefits fraud, the public perception has long been that it is rampant, but sentencing and enforcement data show it remains a small fraction of total cases, even though cases have risen in recent years. For example, in fiscal year 2024, the U.S. Sentencing Commission reports 937 cases of government benefits fraud out of 61,678 total federal cases; that category of offenses has indeed increased markedly since 2020, but it is still numerically small relative to the scale of benefits programs. Policy think‑tank work tends to focus less on “massive recipient fraud” and more on structural design issues in means‑tested programs.[7][8]
The analogy to elections cuts both ways:
- It’s a reminder that sometimes “everyone says there’s tons of fraud” does not match measured reality.[8][7]
- It also supports your instinct that we should not dismiss fraud concerns out of hand; instead, we should look to systematic data, audits, and prosecutions rather than anecdotes.[7][8][5]
## How to evaluate claims like this going forward
When you see a claim like “machines showed 511,000 votes but the state recorded 567,000”:
1. Identify the exact election (which proposition, which date, which county or set of counties).
2. Pull the official canvass results and any public audit/recount reports from the county registrar or secretary of state.
3. See whether independent fact‑checkers or legislative research bodies have reviewed that specific claim.[6][5]
4. Compare: are people citing preliminary machine logs, a subset of voting modes (e.g., in‑person but not mail), or misreading cumulative vs. precinct‑level totals?
Doing that kind of trace‑back is what turned the “45,000 missing ballots” narrative in Riverside into a documented 103‑ballot variance subject to normal reconciliation, not a sign of systemwide fraud.[2][3][1]
So your bottom line—“if there’s a real discrepancy it should be investigated”—is sound, but the particular 511,000 vs 567,000 example, as framed, looks far more like another unvetted numerical claim than a proven red flag in the official records.[3][1][2][5]
Thank you for the information. It gives me a point to start with. Have no problem being proven wrong, truth is my aim, much more than being right. That was a lot to go through, your effort is appreciated.
It's so interesting reading this just as I'm working my way through the section of Doris Kearns Goodwin's biography of LBJ where she talks about him learning the procedures and levers of power in the Senate and using them to run the show. I can't believe actual senators can't be bothered to do this even today.
The Senate has lost it's vibrancy; has become a complacent body full of yes-people and no longer serves it's original purpose. This is not the fault of the Senate, it is the result of an impervious population which votes for a party rather than a person resulting in representatives who care little about issues and vote according to their party's prescription.
More generally- Make America America Again.
There's nothing quite like a bill that seeks to solve a problem that barely exists. Voter fraud is very rare in America, despite the president's claims, which have a chilly relationship with the truth.
The Save America Act is a tool for voter suppression, not election integrity.
What a fabulous piece of work Gabe. Hugely informative, but hugely depressing that they are so coddled into conformity and laziness. While not mentioned in this piece, I surely agree with docking their luxurious pay/privileges until they fund the folks currently working without pay, like the TSA.
Reading the comments from the senators to Gabe was so disheartening. They are so cut off from "normal" America and don't seem to care about how Americans are affected by their actions. But I don't know if there's any way back from this. Anyone who tries to make a difference quickly discovers how hard it is to swim upstream.
Great article, Gabe! Thanks again for some amazing insight! And how cool is that you got to walk around the Senate asking truly interesting questions.
Great work Gabe ! If the Democrats are not being allowed to submit amendments, how can the Senator interviewed really believe they are engaged in a meaningful process. Sounds like: here's the bill like it or not...or I suppose debate could convince Republican Senators to amend the bill, assuming they weren't so scared of the _____________.
This, depressingly, reminded me of what Romney said about being told when he became a senator that only 20 of them do all the work and the other 80 are just along for the ride. 🤦🏻♀️
I keep a copy of the Constitution on my desk. Still, until the Save Act, I thought the requirement that only citizens vote in federal elections was in the Constitution. According to AI (my new best friend), since 1924, the requirement is in either the state constitutions or the statutes of all 50 states. (I know the significance of the 1924 date.) This means to me that states can change that requirement. We also do not have any national registry of citizens. The Save Act to me is an insurance policy. The cost (people will have to locate proof of citizenship) benefit (there is presently little voter fraud) analysis does not resonate with me. A belief in our election integrity is an intangible benefit. I want to confine the benefit to citizens whose first loyalty is to this country and who stand willing to defend it when necessary, not just those who seek the economic benefits.
I hope the conservative contingent does pull off one of those crazy scenarios. Sleeping on the floor and/or standing up and speaking (and not going to the head of the line at airports) would be a good thing.
Great column!
Rosemary, I genuinely used to enjoy your comments. You are clearly different from me. Different part of the country, different politics, and different career (I seem to recall former lawyer?). Although I appreciate your transparency around using AI to contribute to your comments and thinking, I find myself becoming less trustworthy of your comments and thinking.
I use AI every day in my work, and I want to urge you to tread carefully when using it. I’ve already seen other folks push back on your AI informed arguments. It isn’t the tool we think it is!
Sincerely, from someone who wants to hear your POV, not an AI informed or enforced one.
Thank you. I am still in testing phase with Gemini Google. It is a Godsend for a quick jolt to a flawed memory. And, I am pleased when there is pushback. For eg, I will do a little more research into the source of citizen restriction re: voting. (I think Mr. Miller reads the 14th Amendment too broadly.) It appears that the restrictions on non citizens came in conjunction with immigration policy (and are, coincidentally loosening for the same reason.) I belong to a group--braverangels.org--which fosters civil discourse about politics, etc.
I recall a past post of yours about braver angels, and I now keep an eye on what they’re doing!
AI is a complicated tool, we are still learning so much. I look forward your continued commentary.
And much thanks to Gabe for creating an atmosphere of curiosity and enhancement that extends to the readers’ comments.
The 14th Amendment, Section 1, is very clear that citizens who are born in the U.S. or who are naturalized are the only people eligible to vote. No state can change that requirement. And since even republicans concede that voter fraud is almost nonexistent, the only purpose of the SAVE Act is to make it hard for people (mostly women, who tend to vote democratic) to prove their citizenship by having to go to the time, effort and sometimes expense to dig up their birth certificates and wedding records to explain why their last name on their drivers license is now different from what’s on their birth records. It’s called voter suppression and it’s as clever as it is evil. Why do you support it?
I hate it when people argue that "women" or "minorities" will be adversely affected as if the class of women or minorities are too addled to obtain documents to prove citizenship. I reject that premise (as a woman) and I would be equally offended by the assertion if I were a "minority." Are you a White Male?
It would be fun to see if they were willing to whittle the SAVE AMERICA ACT down to the basics, on having proper ID for voting and nothing else. That might be popular enough to actually pass. The polls show 81% for it, with a call in from their constituents, it might pass. Heavy on the might.
But why bother spending ANY time on legislation designed to solve a problem that doesn’t exist? Oh, right, so republicans can suppress the vote while appearing to be concerned about “election security.”
I appreciate your view, but the same was said of welfare fraud, and that seems to be proving less than an accurate statement of fact. In Southern California in a recent statewide referendum, the vote tally on the machines was roughly 511,000+, the vote tally that was recorded was 567,000+. Something is wrong there, and have no evidence of things either way, just that there is a stated discrepancy, which means it should be investigated.
As to voting, the Constitution states it is limited to American citizens, shouldn't they have to prove they are such before voting? I am all for having the state or federal government pay for any effort to find birth certificates or whatever else is needed. The question remains, shouldn't there be some proof of citizenship, before one votes?
I decided to check out your first comment using Perplexity. The result is so long I’m not gonna inquire about your second point! It’s interesting though. Here goes:
The claim you quote about a 511,000 vs 567,000 machine‑count discrepancy in a Southern California referendum is almost certainly based on a misunderstanding or misreporting of the underlying data, and as stated it is not strong evidence of large‑scale fraud.[1][2][3]
## On the specific “511,000 vs 567,000” claim
Several recent California controversies look very similar to what you’re describing: citizen groups or activists have alleged large gaps between ballots “logged” or “machine tallied” and ballots “reported,” which local election officials then showed were based on reading preliminary or partial numbers incorrectly.[2][3][1]
In Riverside County, for example, Sheriff Chad Bianco has claimed a roughly 45,000‑vote discrepancy in a redistricting referendum, prompting him to seize hundreds of thousands of ballots for his own hand count. The county registrar, however, explained that when the actual certified records are compared, the difference between ballots cast and ballots counted is about 103 votes, roughly 0.016%—a level consistent with normal administrative variance (spoiled ballots, provisional ballots rejected, etc.). In other words, the headline “tens of thousands of missing votes” arose from misinterpreting raw logs and preliminary machine data, not a real gap in the certified totals.[3][4][1][2]
Your 511,000 vs 567,000 figures fit that same pattern: a large, round discrepancy claim circulating without a clear link to the official canvass reports, likely mixing apples (e.g., one county’s machine totals at a point in time) with oranges (multi‑county or statewide certified numbers, including late‑counted mail and provisional ballots). Without a specific jurisdiction, contest name, and official canvass, the claim by itself is weak evidence—what we can say from similar, well‑documented disputes is that when election officials and independent reviews dig into the details, the “huge gap” tends to collapse to a tiny one explained by normal election processes.[5][1][2][3]
## Does a discrepancy mean “fraud” or “shouldn’t be investigated”?
You’re right on one core point: if there is a genuine, documented discrepancy between ballots cast and ballots counted, it should be investigated. Election systems expect reconciliation: every batch and machine has logs, and canvass procedures are designed precisely to catch and correct mismatches. A healthy system:[5]
- Tracks differences between machine counts, hand counts, and reported totals.
- Explains them in public reports (spoiled ballots, duplicates, provisional ballots rejected, etc.).
- Escalates to audits or recounts when thresholds are exceeded.
That’s different, though, from treating any large number someone quotes online as proof of fraud. The Riverside case shows how quickly a claim of a 45,000‑vote gap can circulate, only for the documented difference to be on the order of a hundred votes once the actual records are examined. So “there is a stated discrepancy” is not the same thing as “there is a verified discrepancy in the official canvass that survived audit.”[1][2][3]
A reasonable standard is:
- Yes, investigate specific, documented mismatches in official records.
- No, don’t presume fraud based on unresolved or third‑hand numerical claims, especially when similar claims have repeatedly been debunked after basic reconciliation work.[6][1][5]
## Welfare fraud as an analogy
On welfare and government‑benefits fraud, the public perception has long been that it is rampant, but sentencing and enforcement data show it remains a small fraction of total cases, even though cases have risen in recent years. For example, in fiscal year 2024, the U.S. Sentencing Commission reports 937 cases of government benefits fraud out of 61,678 total federal cases; that category of offenses has indeed increased markedly since 2020, but it is still numerically small relative to the scale of benefits programs. Policy think‑tank work tends to focus less on “massive recipient fraud” and more on structural design issues in means‑tested programs.[7][8]
The analogy to elections cuts both ways:
- It’s a reminder that sometimes “everyone says there’s tons of fraud” does not match measured reality.[8][7]
- It also supports your instinct that we should not dismiss fraud concerns out of hand; instead, we should look to systematic data, audits, and prosecutions rather than anecdotes.[7][8][5]
## How to evaluate claims like this going forward
When you see a claim like “machines showed 511,000 votes but the state recorded 567,000”:
1. Identify the exact election (which proposition, which date, which county or set of counties).
2. Pull the official canvass results and any public audit/recount reports from the county registrar or secretary of state.
3. See whether independent fact‑checkers or legislative research bodies have reviewed that specific claim.[6][5]
4. Compare: are people citing preliminary machine logs, a subset of voting modes (e.g., in‑person but not mail), or misreading cumulative vs. precinct‑level totals?
Doing that kind of trace‑back is what turned the “45,000 missing ballots” narrative in Riverside into a documented 103‑ballot variance subject to normal reconciliation, not a sign of systemwide fraud.[2][3][1]
So your bottom line—“if there’s a real discrepancy it should be investigated”—is sound, but the particular 511,000 vs 567,000 example, as framed, looks far more like another unvetted numerical claim than a proven red flag in the official records.[3][1][2][5]
Sources
[1] GOP sheriff seizes 650,000 ballots in challenge to California ... https://www.democracydocket.com/news-alerts/gop-sheriff-seizes-650000-ballots-in-challenge-to-california-redistricting-vote/
[2] Bianco said the alleged discrepancy amounted to about 45800 votes https://www.facebook.com/NBCSanDiego/posts/bianco-said-the-alleged-discrepancy-amounted-to-about-45800-votes-a-difference-e/1362884942552376/
[3] Riverside County sheriff launches ballot investigation over reported ... https://www.youtube.com/watch?v=kaLhv_sVXew
[4] Bianco seizes ballots, probes claimed 45,000-vote gap in Riverside ... https://www.youtube.com/shorts/XmgrOHwuRRM
[5] [PDF] CRB Literature Review on Voter Fraud in California, 2020-2024 https://selc.senate.ca.gov/system/files/2025-08/devin-lavelle-crb-voter-fraud-presentation-senate-elections-2025-08-26-ada.pdf
[6] White House evidence doesn't prove Trump statement ... - PolitiFact https://www.politifact.com/factchecks/2025/nov/05/donald-trump/California-redistrict-Newsom-rigged-Congress/
[7] Government Benefits Fraud | United States Sentencing Commission https://www.ussc.gov/research/quick-facts/government-benefits-fraud
[8] The Biggest Fraud in Welfare | Cato Institute https://www.cato.org/commentary/biggest-fraud-welfare
[9] New video appears to show felony violations in California - YouTube https://www.youtube.com/watch?v=TA5hDOjcrIA
[10] California investigates ballot initiative petition fraud after viral video https://www.reddit.com/r/California_Politics/comments/1rs37na/california_investigates_ballot_initiative/
Thank you for the information. It gives me a point to start with. Have no problem being proven wrong, truth is my aim, much more than being right. That was a lot to go through, your effort is appreciated.
This is one of many excellent reasons to impose term restrictions! Limit Senators to two terms, and Representatives to 4.
Thank you!! Thank you!! Your columns are always informative!! Thank you for your hard work.
Great explanation. I'm saving this for when ( or if) the Senate gets serious about SAVE America. Or gets serious about being the Senate.
It's so interesting reading this just as I'm working my way through the section of Doris Kearns Goodwin's biography of LBJ where she talks about him learning the procedures and levers of power in the Senate and using them to run the show. I can't believe actual senators can't be bothered to do this even today.
Thank God for folks like senator Kennedy.
https://torrancestephensphd.substack.com/p/senator-john-kennedy-a-national-treasure
My book club knows more about the SAVE Act than these folks!! Sad.
The Senate has lost it's vibrancy; has become a complacent body full of yes-people and no longer serves it's original purpose. This is not the fault of the Senate, it is the result of an impervious population which votes for a party rather than a person resulting in representatives who care little about issues and vote according to their party's prescription.