In a bold legal move, Reddit has taken Perplexity to court, accusing the AI startup of unlawfully collecting information from its site to improve its AI-driven search platform. The complaint also targets three additional firms accused of supplying scraping software that enabled unauthorized access.
According to Reddit, Perplexity extracted and reused large portions of its material without any consent, turning the content into answers through its AI “answer engine.” The social network emphasized that it is among the most frequently cited data sources for AI models and insists that all organizations must respect its content usage policies. The company highlighted that major tech players like Google and OpenAI have followed these guidelines appropriately.
In response, Perplexity claimed that its data practices were guided by integrity and responsibility. It maintained that it would continue supporting open access to information, portraying its mission as one that aligns with public benefit and transparency.
The lawsuit also involves Oxylabs from Lithuania, AWMProxy from Russia, and SerpApi based in Texas. Reddit alleges that these entities developed tools that circumvent its security barriers and bypass Google’s built-in protections to retrieve search data. Both Oxylabs and SerpApi have declared their intent to contest the accusations, while AWMProxy has yet to comment publicly.
Documents reveal that Reddit issued a cease-and-desist letter to Perplexity the previous year, demanding an end to the data collection. However, the platform asserts that the AI firm not only ignored the request but allegedly intensified its content use afterward.
Now, Reddit seeks financial compensation and a federal injunction to halt Perplexity from further accessing or exploiting its material. The case represents Reddit’s second major lawsuit concerning AI model training, following a similar complaint it filed against Anthropic earlier in the year.