(
).
Emma Roth is a journalist who covers the streaming battles, consumer technology, crypto, social media and more. She was a writer at MUO.
that racist videos created with Google’s AI-generated video tool Veo 3 received millions of views on TikTok. The AI-generated videos that were uncovered by Media Matters are full of racist tropes and many of them target Black people.
Media Matters concluded that the videos — including one with 14,2 million views — had been created using Google Veo 3 due to the “Veo’ watermark in the corner. Media Matters reported that some users listed hashtags or captions or usernames related to Veo 3 and AI. The organization found clips that lasted no longer than eight seconds or were “composed of several clips, each lasting no longer than eight seconds”, which is in line with Veo 3’s eight-second limitation. Veo 3, launched in May by Google, allows users to create AI video clips and audio using only a text prompt. Google’s website states that it will “block harmful results and requests.” TikTok rulesstate that “hateful behavior and hateful speech has no place on TikTok.”
Ariane de Sellers, TikTok spokesperson, said that “we proactively enforce robust laws against hateful speech, and we have removed the accounts that were identified in the report. Many of them were already banned before the report was published.” The Vergediscovered that some of the videos highlighted in Media Matters had been posted on YouTube, but with less views. Wired discovered that similar AI-generated racist videos were also available on Instagram. Media Matters also found videos with antisemitic and racist content, including depictions of Asians and immigrants.
Google did not immediately respond to The Verge’s request for comment.
Updated, July 2nd. TikTok has added a statement.
