Reddit Inc. sued Perplexity AI Inc. and three other companies over alleged data scraping from the discussion site without permission, a sign of the growing demand and value of original data in the burgeoning AI industry.
Three data scraping companies — Oxylabs UAB, AWMProxy, and SerpApi — have been collecting Reddit data via Google search results for the purpose of reselling it, according to the complaint filed Wednesday in federal court in Manhattan. Perplexity has been buying that data from at least one of the companies, the suit alleges.
Reddit’s growing repository of data has become a valuable commodity given the rise in AI models that rely on massive troves of information for training and surfacing relevant results. Reddit has already inked deals with OpenAI and Alphabet Inc’s Google to license its data for training purposes, but has taken legal action against others it believes to be using the data without a formal agreement.
Reddit sued AI firm Anthropic earlier this year in San Francisco court over similar data scraping allegations.
“AI companies are locked in an arms race for quality human content — and that pressure has fueled an industrial-scale ‘data laundering’ economy,” Ben Lee, Reddit’s chief legal officer, said in a statement shared with Bloomberg News. “Reddit is a prime target because it’s one of the largest and most dynamic collections of human conversation ever created.”
Spokespeople for Perplexity, SerpApi and Oxylabs did not immediately respond to requests for comment. A spokesperson for AWMProxy couldn’t immediately be located.
The case is Reddit Inc. v. SerpApi LLC, 25-cv-08736, US District Court, Southern District of New York (Manhattan).
