Reddit CEO says Microsoft needs to pay to search on the site

After reaching deals with Google and OpenAI, Reddit CEO Steve Huffman is asking Microsoft and others to pay up if they want to continue scraping the site’s data.

“Without these agreements, we don’t have any say or knowledge about how our data is displayed and what it is used for, which has now put us in a position of blocking people who are unwilling to accept how we would like our data to be used.” ​​or not,” Huffman said in an interview this week. He specifically cited Microsoft, Anthropic and Perplexity for refusing to negotiate, saying it has been “a real pain in the ass to block these companies.”

Reddit has been stepping up its fight against trackers in recent months. In early July, its robots.txt file was updated to block web crawlers with which it does not have agreements. Then people started to realize that Reddit results were only visible in Google results – where Reddit is paid to have its data shown – and not in other search engines like Bing.

Huffman said that Microsoft has been using Reddit data to train its AI and summarize its content in Bing results “without telling us,” and that Reddit data has also been sold through the Bing API to other search engines. In the interview, he referenced Microsoft AI CEO Mustafa Suleyman’s recent comment at a conference that public data on the Internet is “freeware.”

“Microsoft, Anthropic and Perplexity acted as if all content on the Internet was free to use,” Huffman said. “That’s their real position.”

In response to Reddit results recently disappearing from Bing, Microsoft search chief Jordi Ribas said on X that “Reddit blocked Bing from crawling its site for search, favoring another search engine and impacting competition from Bing and Bing-powered engines.” Microsoft spokeswoman Caitlin Roulston said separately On the edge last week that “we honor instructions provided by sites that do not wish to have content on their pages used with our generative AI models.”

Huffman pointed to OpenAI’s recent announcement of SearchGPT, which will be able to show results from Reddit thanks to a deal both companies reached earlier this year, as the model he wants to replicate. None of the content licensing deals Reddit has made so far include exclusive use cases for its data, according to spokesman Tim Rathschmidt.

By asking for licensing deals, Reddit is joining more traditional media publishers (including On the edge parent company, Vox Media) in seeking payment for allowing its content to power generative AI. “I think the traditional value exchange of search engines has changed,” Huffman said. “Search, summarization, and training are merging, and the tradeoff of tracking value for return traffic is becoming confusing.”

Spokespeople for Microsoft, Anthropic and Perplexity had no comment for this story by press time.