ChatGPT has been generating New Yorker-style cartoons that include signatures of real cartoonists without their permission. In late August, a fan posted an AI-generated image on Facebook depicting Dolly Parton and Tim Curry in a celestial setting, signed with the pen name “BLOPER” used by New Yorker cartoonist Brendan Loper. Loper discovered the image after a friend shared it, and expressed discomfort that his name appeared on content he did not create. This was not an isolated case. Over the summer, similar AI-generated cartoons bearing the signatures of other New Yorker contributors — including Emily Flake, Pat Byrnes, Jason Adam Katzenstein, and others — circulated across social media platforms and forums such as Reddit. In some instances, the signatures were exact replicas; in others, they appeared as stylized approximations.

Cartoonists have described the unauthorized use of their signatures as a violation of personal identity and professional integrity. Emily Flake noted that seeing her name attached to AI-generated work felt like “seeing a distant cousin” — recognizable, yet disconnected from her actual creative process. Legal experts suggest that while copyright law may not fully address such cases, claims under right-of-publicity statutes could offer recourse if the signatures are used in commercial contexts. The issue has gained attention amid broader concerns about AI systems trained on copyrighted material. OpenAI has acknowledged the reports and stated that it is investigating the matter, though it has not disclosed details about how its models ingested or replicated these signatures. The company previously entered a licensing agreement with Condé Nast in 2024 to use New Yorker content, but that agreement does not cover AI training on cartoonist signatures. As AI-generated imagery becomes more prevalent, questions about attribution, consent, and compensation continue to shape debates over creative labor in the digital age.