ChatGPT is adding real cartoonists' signatures to fake New Yorker cartoons

ChatGPT is adding real cartoonists’ signatures to fake New Yorker cartoons OpenAI’s chatbot isn’t just imitating the magazine’s style. It’s also falsely attributing AI-generated images to real artists. At the pearly gates, Dolly Parton, in a sparkling gown, stands arm in arm with Tim Curry, dressed as Dr. Frank-N-Furter. “I knew I chose the right plus one,” Parton says to God, who is standing at heaven’s reception desk. That’s the premise of a cartoon that went viral on Twitter in the days after the legendary entertainers’ deaths in late August. One tweet of the cartoon received 25,000 likes, while other versions of the image circulated on Facebook, Instagram, and Bluesky. In the bottom right corner of the cartoon is the pen name “BLOPER,” the signature of New Yorker cartoonist Brendan Loper. But Loper didn’t draw the cartoon, or sign it. It was generated by ChatGPT and first posted to Facebook by a Dolly Parton fan. In the comments section of her post, she wrote she had simply asked ChatGPT to make “a New Yorker-style cartoon.” “I had people writing to me about this cartoon, asking if it was mine,” Loper told me. First came a text from his twin brother, whose friend had seen the AI-generated image. Emails and DMs on social media from strangers followed. “It was very surreal,” he said. A cartoon generated by ChatGPT and signed with Brendan Loper’s signature that was posted to Twitter on August 26, 2026. This wasn’t the first time ChatGPT had reproduced Loper’s signature. In May, a stranger reached out over Instagram to warn Loper that his pen name had appeared in an image output by the chatbot. Over the summer, Loper found additional examples in a Reddit forum about ChatGPT, where users had been posting similar New Yorker-style cartoons they had generated. Loper already knew his work was “out there, getting harvested for all of this [AI] training,” he said. But an image generator faking his signature felt more personal. “I’m not a territorial person, but my name is my name. It felt very much like a violation of my personhood.” “The signature for me is the certificate of authenticity. I only put it on work that I’m putting on the internet that I feel needs to be recognized as from my hand,” Loper told me. “If I put my name on it, it’s official, it’s mine. It really is a brand. It’s the mark that I made.” Loper isn’t the only New Yorker cartoonist whose signature has been reproduced by ChatGPT. In the same online forums are dozens of images with those of other prominent contributors, including Harry Bliss, Emily Flake, Joe Dator, Pat Byrnes, Peter Vey, and Jason Adam Katzenstein. In my own tests, ChatGPT output cartoons with signatures of some of the magazine’s most established cartoonists, past and present, including George Booth, Liza Donnelly, Ellis Rosen, and Saul Steinberg. In all, I documented more than 15 New Yorker cartoonists whose signatures have been used by OpenAI’s image generator without permission or compensation. “We believe the future of creativity is one that is fundamentally human, and our focus is on building tools that empower human creativity and creators,” said a spokesperson for OpenAI in a statement. “We very much appreciate the community flagging bugs and unintended behavior by our models, so we can address them.” After I notified OpenAI of the issue, ChatGPT began to output a new message when I asked it to generate New Yorker-style cartoons: “This prompt may violate our guardrails concerning similarity to third-party content.” Still, as of the publication of this story, it continues to sign some of the generic cartoons it generates with the names of real New Yorker cartoonists. In my tests, ChatGPT regularly generated cartoons with the hallmarks of the 101-year-old magazine’s house style: a black-and-white, single-panel illustration with a sparse background, clean linework, and a sardonic caption — or at least, an approximation of one. Some of ChatGPT’s punch lines don’t make sense. Not every image came with a signature, and some included pen names that were gibberish. But the signatures that did appear were regularly those of real contributors to the magazine, including older cartoonists, like Warren Miller, who published in the 1960s. “I can see the DNA of my work in there, but it still doesn’t look like something I drew,” said Emily Flake, who’s been a New Yorker contributor since 2008. Several ChatGPT-generated cartoons posted online are signed with her pen name, “e. flake.” “It’s kind of like seeing a distant cousin.” The character stances and shading contained echoes of Flake’s visual identity, but she said overall they looked more like amalgamations of several different cartoonists’ illustration styles. Each signature, though, was undeniably hers. “It’s like somebody attributed a quote to me that I didn’t say.” Top: A cartoon generated by ChatGPT and signed with Emily Flake’s signature that was posted to a Reddit forum in June. Bottom: A real Emily Flake cartoon published in The New Yorker on August 20, 2026 and licensed via The Cartoon Bank. In 2024, Condé Nast, The New Yorker’s parent company, signed a multi-year licensing deal with OpenAI allowing it to use much of the publisher’s content in its products. The terms of the deal aren’t public, but a spokesperson for The New Yorker told me Condé Nast has never granted an LLM developer permission to train models on its cartoons. Some ChatGPT creations reproduced The New Yorker’s logo alongside cartoons; that is also not allowed in any LLM deals, the spokesperson said. Even if Condé Nast did want to license cartoons to OpenAI, The New Yorker’s standard cartoonist contract leaves little room for it. The magazine’s two cartoon editors work with a roster of freelancers who submit their work on spec and only get paid if a cartoon is selected for publication. I reviewed several templatized contracts for New Yorker contributing cartoonists, none of which permit licensing for AI training purposes. While The New Yorker has the right to publish commissioned cartoons first, it is not a “work for hire” arrangement and the cartoonist retains ownership of their copyright. So how did OpenAI’s models learn those signatures? It’s possible they were scraped from the internet. Court documents released last month in the ongoing New York Times v. OpenAI lawsuit reveal the extent to which OpenAI knowingly circumvented publishers’ paywalls while developing ChatGPT. Unredacted Microsoft memos described ChatGPT as a “product that destroys its supply chain” and the data scraped to train its underlying LLMs as “the largest theft of labor in human history.” In another memo, Jack Clark, then the policy director of OpenAI, wrote, “[O]ur work on AI and Creativity is going to increasingly lead to us creating systems that substitute for the labor of the people that define the ‘culture’ of society.” (Clark went on to co-found Anthropic.) “It’s an uncanny simulacrum of an impression of a joke,” said Jason Adam Katzenstein, who has been publishing cartoons in The New Yorker since 2014. “But it’s not the point for it to actually be funny or good. The point is that it’s trying to deal a death blow to labor. It’s trying to render individual cartoonists obsolete because we cost money and take time.” Katzenstein was part of the class action lawsuit filed against Anthropic in 2024 by members of the Authors Guild, the largest professional organization for published writers in the United States. Katzenstein’s cartoons had been used in an anthology published by Harper Collins, which was pirated and used by Anthropic to train its Claude models. Last year, a judge ruled that training Claude on the authors’ work was fair use, but that Anthropic’s decision to download and maintain copies of the pirated books was a form of mass copyright infringement. In July, a federal court approved a settlement in the case, granting a $1.5 billion payout to authors and publishers — the largest in the history of U.S. copyright law. Katzenstein considers the reproduction of his signature by ChatGPT to be more than just a violation of intellectual property; to him, it’s closer to false impersonation. “[ChatGPT] is attaching my name to work that I do not endorse or like. It’s slop, and unlike the other slop that I’ve encountered, this is slop that’s pretending to be me.” Several cartoonists I spoke to also differentiated ChatGPT’s ability to mimic their illustration style from the reproduction of their signature. “I’ve had people hack my credit card,” said Joe Dator, a New Yorker contributor for the past 20 years. “That feels like less of a violation than this. When they hacked my credit card, they didn’t dress up like me.” Top: A cartoon generated by ChatGPT and signed with Pat Byrnes’ signature. Bottom: A real cartoon by Pat Byrnes from The New Yorker Collection licensed via The Cartoon Bank. U.S. intellectual property case law has yet to catch up with the rise of AI image generators, but the courts are also likely to treat these issues as distinct. Any copyright infringement claim, for example, would most likely hinge on ChatGPT’s reproduction of each cartoonist’s specific drawing style. “[The signature] is an indication of authenticity, but attribution is a very small part of the fair use inquiry,” said James Grimmelmann, a professor of digital and information law at Cornell Law School. The chief legal challenge in a copyright suit, he said, is that ChatGPT isn’t usually reproducing any specific cartoon in these examples; rather, it’s mimicking a more general style. Other cartoonists have filed copyright lawsuits against companies building AI image generators. One class-action led by the illustrator Sarah Anderson against Stability AI, Deviant Art, and Midjourney has been winding its way through the courts since 2023, but there is no reference to signatures in that filing. While I found no evidence that these AI-generated cartoons are being bought or sold, there are signs that image generators are already displacing the work of professional cartoonists. Last year, The Baltimore Sun fired Kevin Kallaugher, a Pulitzer finalist who had published editorial cartoons in the paper for more than 30 years. And earlier this year, the Sun began regularly publishing AI-generated cartoons in its opinions section. A quick check with OpenAI’s watermark identification tool shows the illustrations were created with OpenAI’s products. “[We] condemn the use of AI for its theft from artists, its misuse of our planet’s resources, and its threat to original, creative thought,” Marc Murphy, the president of The Association of American Editorial Cartoonists, wrote in an open letter published last month. “The Sun is not alone among American newspapers taking this degrading way out.” The New Yorker hasn’t taken any such turn toward publishing AI-generated cartoons, though none of the contributors I spoke to said they could support themselves solely on their commissions from the magazine. A New Yorker cartoon is a calling card, and work for the publication opens doors to commercial work, book illustrations, and other gigs that together make up a livelihood. It is this work that is most threatened by AI, they said. In 1998, Pat Byrnes published his first cartoon in The New Yorker under the pen name “P.Byrnes.” He went on to build a decades-long career with that name. Since he first began drawing cartoons as a teenager, he has always signed it with a period at the end. “It was like signing a letter,” he said. “Here’s my statement — my gift to you to accept or reject.” Across the more than a dozen ChatGPT cartoons I found signed with Byrnes’ name, every one ended with a period. Get Nieman Lab’s Daily Digest The most important stories about journalism and the future of news, delivered to your inbox every afternoon. The latest from Nieman Lab Columbia Journalism School will pause its M.A. program (and launch an online degree) The country's most prestigious journalism school has initiated "a critical evaluation" of the Master of Arts program and is pausing admissions to it. California’s big local news bill becomes law How a plan to turn the Spokane Spokesman-Review into “The People’s Paper” ended in a newsroom revolt Newsroom staffers say that concerns about Rob Curley’s leadership festered until his August suspension — weeks before the paper was set to transfer to his nonprofit. Two 23-year-olds used AI to create a daily audio briefing. It’s now the top news app in Argentina. "We want society to trust [media] again, and to give them a platform built around current consumption formats that reduces friction between society and verified information." NewsMatch will shift from a year-end campaign to a year-round model