TikTok is seeing an influx of AI-generated videos showing racist stereotypes. The videos all appear to be generated by Google’s recently launched Veo 3 tool, which can produce detailed AI videos.
Spotted by non-profit watchdog Media Matters, a new report shares examples of many racist clips and how it believes Google’s Veo AI tool has been involved in the production. The videos were all uploaded to TikTok as individual short clips, but Media Matters has put together many of clips on YouTube with further context about each one.
Many of the racist stereotypes depicted are negative toward Black people in the United States. Some of the videos show monkeys, an insult often used by racists, as well as showing Black people as absent parents or criminals.
Other videos include antisemitic tropes, while some clips specifically target immigrants. Media Matters also found clips online referring to historical traumas, such as the Holocaust and the Ku Klux Klan’s attacks on Black Americans. Be aware that the following video below includes racist content readers may find distressing.
Each of the clips are designed to outrage users, encourage reactions, and therefore reach more people through TikTok’s algorithm.
Exactly how these clips were produced remains unclear, but evidence points toward Google’s new Veo 3. The clips are all eight seconds long, which is the current time limit on video production in Veo 3. Some of the clips also have the word Veo in the bottom right corner.
Media Matters also found there were multiple references to Veo 3 across the clips with some using it within the captions or hashtags, while some had Veo included in the account’s username. Many of the clips also have mistakes associated with AI within the captions.
A spokesperson for TikTok told The Verge, “We proactively enforce robust rules against hateful speech and behavior and have removed the accounts we identified in the report, many of which were already banned prior to the report publishing.”
Recommended by Our Editors
TikTok’s community guidelines say, “We do not allow any hate speech, hateful behavior, or promotion of hateful ideologies. This includes explicit or implicit content that attacks a protected group.”
However, these videos aren’t just limited to TikTok. The Verge found some examples of the videos being viewable on YouTube, while different racist AI-generated videos were found on Instagram by Wired.
AI-generated hate speech may be the next big content moderation battleground for the world’s biggest social networks and video hosting platforms. If you spot videos that you believe show racist stereotypes or other harmful content, be sure to use that platform’s reporting tools to highlight issues.
Get Our Best Stories!
Your Daily Dose of Our Top Tech News
By clicking Sign Me Up, you confirm you are 16+ and agree to our Terms of Use and Privacy Policy.
Thanks for signing up!
Your subscription has been confirmed. Keep an eye on your inbox!
About James Peckham
Reporter
