ChatGPT gains Yelp reviews, ratings for local answers

OpenAI has licensed Yelp’s reviews, ratings, photos and business information so ChatGPT can answer local questions with that data. Yelp first disclosed the agreement in its February 2026 results, and the operational scope emerged this week when Axios reported how the two companies plan to use it. For any brand with physical locations, it means the reviews and details attached to those locations now feed a mainstream AI assistant, not just a search engine, and the accuracy of that data decides how the brand is described when a customer asks ChatGPT where to go.

What happened

Yelp said in its full-year 2025 earnings release, published on 12 February 2026, that it had signed an agreement with OpenAI. The details became clearer on 23 July 2026, when Axios reported that ChatGPT can surface Yelp reviews, star ratings, photos and business information inside its answers to local queries, with Yelp branding and links shown alongside the content, though OpenAI controls how that appears. Axios reported that the arrangement is non-exclusive and that Yelp’s Request a Quote flow is set to follow, letting people contact local service providers from inside the chat. Yelp reports its network holds around 330 million cumulative reviews and more than 8 million business listings. The financial terms were not disclosed.

Why it matters

The deal is a clear signal that AI assistants are becoming a local discovery surface in their own right, licensing structured review and business data rather than only scraping the open web. That reframes what brand visibility means: it is no longer only about ranking in Google or appearing in AI Overviews, but about whether the data an assistant draws on describes each location correctly. Reviews in particular are now dual-purpose, shaping both a customer’s read of a location and the summary an AI gives when asked to recommend one.

Yelp reports its network holds around 330 million cumulative reviews and more than 8 million business listings.

Yelp

What this means for multi-location brands

For a central team, the takeaway is not to court one data provider but to make every location’s own data unmistakable wherever assistants read it. Keep names, addresses, categories and hours consistent across the estate from a single local business listing source of truth, because an assistant that pulls a stale hour or a wrong category will repeat it confidently to a customer. Treat review management as visibility work as much as reputation work: keeping reviews current and answered across every location protects both the customer’s impression and the raw material these assistants now summarise. And plan for the assistants directly, optimising the estate for generative engine optimization rather than assuming a strong Google Business Profile presence carries over automatically to ChatGPT, Gemini and the rest.

The bottom line

Licensing deals like this one will decide which brands AI assistants can describe accurately and which they get wrong. The brands that win are the ones whose location data and reviews are already governed centrally, so whatever source an assistant licenses next, it finds the same correct answer.