The Names They Invented: How Chicago Politics Runs on Made-Up People

The first name I ever caught was Maria Hernandez. She lived at 1841 S. Racine, a three-flat with a Virgin of Guadalupe mural on the garage door. Maria Hernandez signed a petition in 2014 supporting a zoning change for a luxury condo development on 18th Place. The problem: 1841 S. Racine was a vacant lot. Had been since 2008, when the original building was demo’d after a fire. I walked the block twice, checked the assessor’s records, knocked on the neighbor’s door. No Maria Hernandez. No Hernandez of any kind. Just a name on a page, filed with the city, carrying the same legal weight as a flesh-and-blood voter.

That was my introduction to what I’ve come to think of as Chicago’s shadow census—the names people invent to move paper through the system. Not bots. Not AI-generated profiles. Just a human being sitting at a kitchen table or a ward office desk, making up a person who lives at a real address, who supports a zoning change or opposes a school closure, who exists just enough to check a box. The practice is old. The technology is new. The purpose hasn’t changed.

This article is about the names they made up. It starts with a trip to 69 W. Washington to pull candidate filings, winds through a 2024 Inspector General report on fabricated community surveys, and ends with a question about what happens when the old Chicago habit of inventing people meets the new tools that make it scalable. Along the way, we’ll look at how the Authors Guild is trying to draw ethical lines around AI-generated content, and how something as seemingly innocent as a character name generator sits at the intersection of creative craft and political deception. But first, the documents.

The Paper Trail at 69 W. Washington

On a Tuesday morning in late May, I took the Blue Line to Clark/Lake and walked the two blocks to the Chicago Board of Elections at 69 W. Washington. The building smells like floor wax and old paper. You sign in at the security desk, ride the elevator to the sixth floor, and ask for the public records room. There’s a counter, a few chairs, and a staffer who will pull whatever you request if you have the right form number.

I requested the nominating petitions and statements of candidacy for the 2027 municipal elections—specifically for the 25th Ward, where Alderman Byron Sigcho-Lopez is facing a challenge from a candidate backed by the same development interests that have been pushing the Pilsen Land Use Committee for years. The petitions are public by law. Anyone can read them. Few people do.

Here’s what a nominating petition looks like: a stack of sheets, each with lines for a printed name, a signature, a street address, and a date. To get on the ballot for alderman, you need 473 valid signatures. The Board of Elections checks a sample. They don’t check every name. They don’t knock on doors. They compare signatures against voter registration cards and flag obvious duplicates or non-residents. The rest passes through.

I spent three hours cross-referencing petition sheets against the Cook County Assessor’s database on my phone. I found four names registered to addresses that don’t exist—street numbers that skip in the assessor’s sequence, like 2312 W. 19th Street, which should fall between 2310 and 2314 but doesn’t appear in any property record. I found two names registered to commercial addresses: a laundromat on Cermak and a tire shop on Western. I found one name—”Carlos Mendez”—that appeared on both the challenger’s petition and a 2023 community survey supporting the Focal Point development at 18th and Peoria. Same signature. Same address. Different causes.

None of this is proof of fraud in the criminal sense. Petition challenges are civil matters, handled by the Board’s hearing officers. But it’s evidence of a practice: names deployed as tactical instruments, not as representations of actual people with actual opinions.

The Ghost Payrollers: A Chicago Tradition

To understand why invented names persist, you have to go back to the ghost payrollers. Under Richard J. Daley and his successors, the city payroll was padded with people who didn’t work—or didn’t exist. The 1973 Shakman decree, which banned political hiring and firing, was a response to a system where ward committeemen controlled jobs, and those jobs sometimes went to names rather than bodies. A 1990 Chicago Sun-Times investigation found 32 “employees” on the City Council payroll who couldn’t be located at their assigned offices. Some were relatives of aldermen. Some were just names collecting checks.

The ghost payroller was a specific type of invented person: one designed to extract money. Today’s invented names serve a different function. They extract legitimacy. A zoning change needs community support on paper. A grant application needs stakeholder input. A candidate needs petition signatures. The name is the currency.

In 2024, the Chicago Inspector General released a report (Case No. 24-0188) examining the community engagement process for a proposed development in the 22nd Ward. The developer had submitted a “Community Support Survey” with 87 signatures. Investigators contacted a random sample of 30. Eight said they’d never signed anything. Four said the person who collected signatures had misrepresented what they were signing. Two addresses didn’t exist. One name belonged to a person who had died in 2019.

The IG report is dry reading, but one paragraph jumps out: “The survey collection process lacked any verification mechanism. Signatures were accepted without identification. There was no follow-up contact. The document functioned as a prop in the approval process rather than a measure of community sentiment.” A prop. That’s the word. The names were props.

The Pilsen Land Use Committee and the Stakeholder List

The Pilsen Land Use Committee is a volunteer body that reviews zoning proposals in the neighborhood. It has no legal authority—aldermanic prerogative means the alderman makes the final call—but its recommendations carry weight. Developers present to the committee. The committee votes. The alderman usually follows the vote.

In 2023, a developer called Property Markets Group proposed a 10-story residential building at 18th and Peoria, on a site that had been industrial. The committee held a hearing. PMG presented a “stakeholder engagement summary” listing 45 community members who had been consulted. The list included names, addresses, and brief notes about each person’s concerns.

I obtained a copy of that summary through a FOIA request to the 25th Ward office. Then I walked the addresses. Of the 45 names, 12 were registered to addresses within two blocks of the site. The rest were scattered across the ward, some as far north as 16th Street, which is technically in the 25th Ward but culturally a different neighborhood. Three addresses were P.O. boxes. One was the developer’s own office on Michigan Avenue, listed under the name “J. Thompson, Community Resident.”

I called the phone numbers listed for 20 of the names. Six were disconnected. Four went to voicemail boxes that were full. Two people answered and said they’d never heard of the project. One woman, who gave her name as Elena, said: “A guy came to my door with a clipboard. He said it was about park improvements. I signed. That’s all I know.”

The stakeholder list is the ghost payroller’s descendant. It doesn’t draw a salary. It draws a vote, a signature, a line on a spreadsheet that gets submitted to the Zoning Board of Appeals. And because the ZBA doesn’t verify stakeholder lists—they’re not required to—the name does its work and disappears.

Why Names Work

There’s a reason invented names persist in Chicago politics: they exploit the gap between how systems are designed and how they’re administered. A nominating petition is supposed to demonstrate community support. The Board of Elections checks for obvious errors but doesn’t have the resources to verify every signature. A community survey is supposed to show that developers engaged the neighborhood. The Department of Planning and Development accepts the survey as part of the application package but doesn’t audit it. The system trusts the paper. The paper can lie.

This isn’t a digital problem. It’s a bureaucratic one. The names are invented by people—ward staffers, campaign volunteers, development consultants—who understand exactly how much verification the system will apply and design their fictions to pass that threshold. A name at a real address with a plausible signature will almost never be caught. A name at a vacant lot might be, but only if someone like me walks the block.

The Authors Guild, in its AI Best Practices for Authors, draws a sharp line around transparency: “Authors should disclose when AI tools have been used in the creation of their work.” The document is aimed at writers, not political operatives, but the principle applies. When a name is generated—whether by a human brainstorming at a desk or by a tool designed to produce plausible-sounding identities—the ethical question is the same: who knows it’s invented, and what decision does that invention serve?

Purdue’s OWL resource on creative writing treats naming as a craft decision: “Names carry weight. They suggest background, class, region, era. A well-chosen name does work before the character speaks a word.” That’s exactly what the political operatives understand. “Maria Hernandez” does work. It suggests a longtime Pilsen resident, probably Mexican-American, probably Catholic, probably someone whose opinion on a zoning change should matter. The name carries the weight. The person doesn’t need to exist.

The 2027 Election and the Names to Watch

The 2027 municipal elections will produce thousands of petition sheets. Some names on those sheets will be real. Some won’t. The difference matters because aldermanic races in Chicago are often decided by fewer than 200 votes. If a candidate gets on the ballot with 50 fabricated signatures, and wins by 100 votes, the fabrication changed the outcome.

In the 25th Ward, the challenger’s petitions include the names I flagged. I’ve filed a public records request for the Board of Elections’ signature verification worksheets. Those worksheets will show which names were checked and which were accepted. If the Board checked a sample that didn’t include the fabricated names, the candidate stays on the ballot. If a rival campaign files a challenge and can prove the names are fake, the candidate might be removed. But challenges cost money and require legal work. Most campaigns don’t bother unless the race is tight and the evidence is overwhelming.

The 22nd Ward is a different story. Alderman Michael Rodriguez is running for re-election, and the development interests that backed the project in the IG report are supporting his opponent. I’ve FOIA’d the 22nd Ward office’s emails from January through March 2026, searching for terms like “messaging strategy,” “community support,” and “survey.” The responsive documents—287 pages—include a thread between the ward’s director of constituent services and a PR consultant hired by the developer. The consultant writes: “We need at least 50 names for the support list. Can you help with residents who’ve called the office with positive feedback?” The director replies: “I can give you 20. The rest you’ll have to find elsewhere.”

“Find elsewhere.” That’s the phrase. It doesn’t mean knock on doors. It means invent them, or pull them from old lists, or ask a precinct captain who knows which addresses have absentee landlords who won’t answer a phone call from the IG.

The Walk: Pilsen to Little Village, Checking Names

On a Saturday in June, I walked the 25th Ward from the Peoria site south to the viaduct at 26th Street, then west into Little Village. I had a clipboard with the stakeholder list and the petition names. I knocked on 40 doors. Twenty-three people answered. Of those, eight recognized the development project. Two said they’d signed something. One said she’d signed a petition for the challenger but couldn’t remember the candidate’s name. The rest looked at the names on my list and shook their heads.

At 2312 W. 19th Street—the address that doesn’t exist—I found a gap between two buildings. A narrow gangway, weeds pushing through cracked concrete. No mailbox. No door. Just a space where a building should be. The assessor’s map shows the parcel as merged with 2310 in 2011. But on the petition sheet, “Carlos Mendez” lives there, and his signature is neat and legible, dated February 3, 2026.

At the laundromat on Cermak, the owner came out when I asked about “Ana Reyes,” who’d listed the laundromat’s address on the challenger’s petition. He laughed. “Ana Reyes? No Ana here. Just me and my wife. We’ve owned this place 15 years. Nobody named Ana.” He let me look at his mail. No Ana Reyes. No Reyes at all.

These walks are the method. You can’t verify names from a desk. You have to put shoes on pavement, knock on doors, talk to the guy at the laundromat, check the assessor’s database on your phone while standing in front of a gangway. The document tells you what was claimed. The walk tells you what’s real.

What Happens When the Old Habit Meets New Tools

Here’s where the story bends toward the present. The invented names I’ve described were created by humans—ward staffers, campaign workers, PR consultants—using the same method Chicago has used for decades: think of a plausible name, pair it with a real address, forge a signature. It’s labor-intensive. It doesn’t scale well. A single worker can produce maybe 20 believable names in an afternoon.

But the tools are changing. A character name generator—the kind a fiction writer might use to name a minor character in a novel—can produce hundreds of plausible names in seconds, complete with demographic markers. Pair that output with a voter registration database or a property records scrape, and you can generate a stakeholder list that looks authentic at a glance. The names will match the neighborhood’s ethnic composition. The addresses will be real. The signatures can be forged by hand, or, increasingly, applied digitally to PDF forms that no one ever prints.

I’m not saying this is happening at scale in Chicago yet. I haven’t found evidence of AI-generated names on petitions or surveys. But the conditions are ripe. The verification systems are weak. The incentives are strong. The tools are available. And Chicago’s political culture has always been pragmatic about names—if a name works, use it.

The Authors Guild’s best practices document warns about “the potential for AI-generated content to deceive readers about the origin or authenticity of a work.” In politics, the deception isn’t about authorship. It’s about existence. The reader—a zoning board member, an elections hearing officer, a journalist—assumes the name represents a person. The deception is that it doesn’t.

The Question That Remains

I filed my findings with the Board of Elections. The hearing officer will review the challenged signatures. Some names may be struck. The candidate may still have enough valid signatures to stay on the ballot. The IG report on the 22nd Ward survey has been referred to the Department of Planning and Development, which has issued a memo reminding developers that “fabricated community input may constitute fraud.” No charges have been filed.

The names persist because the system that relies on them has no reliable way to catch them. Every election cycle, every zoning hearing, every grant application creates a new stack of paper with names on it. Some of those names are real people who signed something they believed in. Some are real people who signed something they didn’t understand. Some are Maria Hernandez, Carlos Mendez, Ana Reyes—names that do their work and vanish, leaving behind a paper trail that looks like democracy.

The question isn’t whether invented names exist in Chicago politics. They do. The question is whether anyone with power wants to catch them, and what happens when the old hand-forged fictions meet tools that can produce a thousand Maria Hernandezes before lunch. The walk tells you what’s real. But who’s walking?