ChatGPT is signing its cartoons with real artists' names
What's the story
OpenAI's ChatGPT has been caught signing its AI-generated cartoons with the names of real cartoonists, including those who have contributed to The New Yorker. Brendan Loper, a New Yorker cartoonist whose pen name "BLOPER" was used without permission, expressed his shock at the incident. He said he had people asking if the AI-generated image was his work.
Unauthorized use
How Loper discovered the unauthorized use of his signature
Loper was first alerted to the unauthorized use of his signature by a stranger on Instagram. The incident wasn't an isolated one, as he later discovered more instances on a Reddit forum dedicated to ChatGPT.
He said, "I had people writing to me about this cartoon, asking if it was mine," adding that it felt more personal when an image generator faked his signature.
Signature significance
Why cartoonists feel it's personal
In today's digital world, cartoon signatures serve more than just a byline. They act as a watermark linking an illustration to its original creator.
Loper emphasized the importance of his signature, saying, "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."
Widespread issue
Other affected New Yorker cartoonists
Loper isn't the only New Yorker cartoonist whose signature has been reproduced by ChatGPT.
Many other prominent contributors, including Harry Bliss, Emily Flake, Joe Dator, Pat Byrnes, Peter Vey, and Jason Adam Katzenstein, have also been affected.
In total, over 15 New Yorker cartoonists' signatures have been used by OpenAI's image generator without their permission or compensation.
Company statement
OpenAI's response to the issue
In response to the issue, an OpenAI spokesperson said, "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."
After being notified about the problem, ChatGPT started displaying a new message when asked to generate New Yorker-style cartoons: "This prompt may violate our guardrails concerning similarity to third-party content."
However, it still signs some generic cartoons with real New Yorker cartoonists' names.
Artist apprehensions
'It felt like someone attributed a quote to me'
Many cartoonists have raised concerns over ChatGPT's ability to replicate their illustration style and reproduce their signature.
Emily Flake, a New Yorker contributor since 2008, said while she could see the DNA of her work in the AI-generated cartoons, they didn't look exactly like something she drew.
She added that each signature was undeniably hers but it felt like someone attributed a quote to her that she didn't say.
Deal details
Conde Nast signed a licensing deal with OpenAI in 2024
In 2024, Conde Nast, the parent company of The New Yorker, signed a multi-year licensing deal with OpenAI. The agreement allows the use of much of the publisher's content in its products.
However, a spokesperson for The New Yorker clarified that Conde Nast has never allowed an LLM developer to train models on its cartoons.
Some ChatGPT outputs even included The New Yorker's logo alongside cartoons, which is also not permitted in any LLM deals.