ChatGPT forges signatures of over 15 New Yorker cartoonists

The case surfaced through a post on X in August 2026, where an image captioned as a ChatGPT-made cartoon carried the signature "BLoper." Brendan Loper told Nieman Jounalism Lab that he received several emails asking whether the cartoon was his, and that he had already seen his signature attached to ChatGPT-generated New Yorker-style cartoons on Instagram and Reddit before. He said he knew his work had been used in AI training, but that faking his signature felt like a more personal violation.

The report places this against a shift in how images circulate. Cartoons and illustrations are increasingly cropped and reposted on social media or embedded in blogs, which makes the signature function as a kind of watermark identifying the original author. Loper described his signature as a certificate of authenticity, applied only to works he wants recognized as his own.

OpenAI's spokesperson told Nieman Jounalism Lab that the company believes the future of creativity is fundamentally human-centered and that it is focused on tools supporting human creativity and creators, and thanked the community for reporting bugs and unintended behavior. Jason Adam Katzenstein, another New Yorker contributor, framed the stakes differently, arguing that AI is threatening cartoonists' livelihoods because their work is costly and slow to produce. How far OpenAI's added warning message changes the behavior of its model is likely to be the test that matters most for the cartoonists involved.