Indie game publisher says TikTok ran racist AI ads without its consent

Finji, the independent game studio behind Night in the Woods and Tunic, has accused TikTok of using generative AI to alter its advertisements without permission, including one that introduced a racist, sexualized stereotype of one of the studio's characters. TikTok declined to comment on the record when contacted by IGN, which first reported the story.ign

The dispute centers on ads Finji runs through TikTok for its games, including the upcoming title Usual June. According to Finji CEO and co-founder Rebekah Saltsman, the studio's TikTok ad account has all AI features disabled, yet AI-modified versions of its ads have been appearing on the platform as if posted directly from Finji's official account.IGN+1

How the Problem Surfaced

Finji first learned of the issue from its own audience. Users commenting on the studio's legitimate ads expressed concern about what they were seeing, and members of Finji's Discord server began sharing screenshots of AI-altered versions. The modified ads appeared to be slideshows rather than the original video format, with at least one image containing a noticeably altered depiction of a Finji character that invoked racial and sexual stereotypes.ign

Saltsman initially raised the alarm on Bluesky, writing, "If you happen to see any Finji ads that look distinctly UN-Finji-like, send me a screencap". She told IGN the studio has AI "turned all the way off" on its TikTok account, and shared backend screenshots confirming that both TikTok's "Smart Creative" and "Automate Creative" features — tools that use generative AI to remix and optimize ad content — were disabled.IGN+1

A Frustrating Support Experience

What followed was a weeks-long struggle with TikTok's customer support. IGN reviewed conversations between Finji and TikTok beginning on February 3, in which a support agent confirmed that Finji's AI settings were off but could not explain how the modified ads were generated. On February 6, TikTok told Finji there was "no evidence that AI-generated content or auto-assembled slideshow assets were added by the system," even after Finji re-sent screenshots of the offending ad.ign

By February 10, after repeated requests to speak with a senior representative, Finji received a response that acknowledged "AI-generated or automatically created content" in its ads and offered to request placement on an "opt-out blocklist" — for which approval was not guaranteed. When Finji pressed further, TikTok said the person responding was already "the highest internal team available for this type of issue" and that the escalation team had "already provided their final findings". As of February 17, Finji had received no further communication.ign

A Broader Pattern

Finji's experience appears to be part of a wider issue on TikTok's ad platform. Other advertisers on Reddit have reported similar problems with AI-altered versions of their ads appearing despite having automated creative features turned off. In August 2025, game developer Nexon faced backlash after AI-generated ads — including deepfakes of real content creators — appeared on the official TikTok account for The First Descendant, with TikTok confirming at the time that it had no system to verify potential copyright infringement in such cases.gamespot+2

"I have to admit I am a bit shocked by TikTok's complete lack of appropriate response to the mess they made," Saltsman said in a statement. "It's one thing to have an algorithm that's racist and sexist, and another thing to use AI to churn content of your paying business partners, and another thing to do it against their consent, and then to also NOT respond to any of those mistakes in a coherent way?"ign