Yelp is going to use artificial intelligence to stitch together user-posted content about restaurants, food, and nightlife to create short videos about these places. The company started testing AI-generated videos last year, but they are now available nationwide on the iOS app’s main feed, which scrolls vertically like TikTok.
Currently, business operators cannot see the videos created for users, and Yelp users also cannot opt out of having their photos or videos appear in AI-stitched Yelp videos. Yelp relies on several generative AI tools to create the finished product: OpenAI LLM writes text descriptions and narrator scripts, compiles story topics, and proofreads, while ElevenLabs is used to generate narrator voices and Amazon Transcribe creates synchronized on-screen captions.
You can get an idea of what they look like in the video below shared by Yelp. The vertical video combines video and images with an AI-generated narrator and AI-generated captions to talk about things like the food, cocktails, and atmosphere of the restaurant.
Yelp wants to do “as many videos as possible,” says Yelp COO Craig Saldanha in an interview with The Verge, but will only do them if the restaurant has enough reviews, photos, and videos to tell a compelling story. Yelp relies on personalized signals to determine when to show you videos. The videos themselves are not personalized, even if they are updated over time – according to Saldanha, there is only one active AI-stitched video about one business at a time live. If a user or business believes that an AI-edited video is inaccurate or offensive, Saldanha says they can report it by clicking on the three dots in the upper right corner of the video. Yelp also conducts periodic “at scale” checks.
AI stitched videos follow other AI-focused Yelp features such as review summaries and review filters.
