Reddit has filed a federal lawsuit against Perplexity AI, a rising star in the AI-driven search space, accusing it of illicitly harvesting user-generated discussions from Google’s search results. The case, brought before the U.S. District Court for the Northern District of California, also targets data firms Oxylabs UAB, AWMProxy, and SerpApi, alleging they conspired to bypass anti-scraping protections on both Reddit and Google to fuel Perplexity’s “answer engine.”
The dispute shines a spotlight on the ethical and legal boundaries of AI’s hunger for data. Reddit, a platform where millions share insights across diverse communities—from parenting forums to tech debates—claims the defendants used sophisticated tactics to skirt its safeguards, including IP limits and CAPTCHA checks. The suit details how the group allegedly employed rotating IP addresses, spoofed device signatures, and server farms to mimic human users, extracting nearly three billion Reddit-containing search results in a two-week span last July. To prove the breach, Reddit’s engineers planted unique, unpublished content in Google’s indexes, only to find it regurgitated hours later in Perplexity’s AI-generated responses.“
These are deliberate violations that erode the trust of our users and the value of our platform,” said Ben Lee, Reddit’s chief legal officer. Having gone public in 2024 and inked data-licensing deals with companies like Google and OpenAI, Reddit argues that Perplexity’s actions undermine its ability to protect user privacy and monetize its content ethically.
Perplexity, a San Francisco startup valued at over $1 billion, markets itself as a conversational alternative to traditional search, synthesizing web data into concise, sourced answers. Backed by investors like Jeff Bezos and NVIDIA, it has drawn praise for its user-friendly interface. But Reddit’s suit paints a darker picture, accusing Perplexity of buying scraped data to sidestep direct licensing agreements. The startup fired back, with spokesperson Jesse Dwyer calling the lawsuit “an attempt to monopolize public data.” In a Reddit thread, Perplexity insisted it processes only publicly available links, comparing its role to a journalist citing sources.
The co-defendants also pushed back. Oxylabs, a Lithuanian data extraction firm, defended its tools as vital for research, from tracking misinformation to environmental studies. SerpApi, based in Texas, cited First Amendment protections, while AWMProxy did not respond. The case, invoking the Digital Millennium Copyright Act and claims of unfair competition, seeks to halt the scraping, destroy collected data, and award damages.
This clash echoes broader tensions in the AI boom, reminiscent of the New York Times’ 2023 suit against OpenAI over unauthorized content use. For Reddit, backed by Advance Publications (parent of Condé Nast), the stakes are high: Its user-driven ecosystem thrives on openness but risks exploitation without clear boundaries. “When data is taken without consent, it’s not just a business issue—it’s about respecting the voices of our communities,” said Dr. Aisha Patel, a digital ethics scholar at Stanford.
As the case unfolds, it could redefine how AI firms navigate the open web, balancing innovation with the rights of content creators. For now, it’s a stark reminder that behind every AI answer lies a human contribution—and a question of who controls its use.
