Encyclopædia Britannica sues Perplexity for copyright infringement

Encyclopædia Britannica and Merriam-Webster have filed a federal lawsuit against artificial intelligence startup Perplexity AI, alleging widespread copyright infringement and trademark violations. The complaint, filed in the Southern District of New York on September 10, 2025, marks another significant legal challenge facing AI companies over their use of copyrighted content.

According to the lawsuit, Perplexity operates what it calls an "answer engine" that "eliminates users' clicks on Plaintiffs' and other web publishers' websites—and, in turn, starves web publishers of revenue—by generating responses to users' queries that substitute the content from other information websites." The publishers claim Perplexity engages in "massive copying" of their protected content without authorization or compensation.

Britannica, founded in 1768, operates as a global digital education platform serving over 150 million students across more than 150 countries. The company recorded over one billion sessions at britannica.com in 2024 alone, excluding an additional 1.4 billion sessions across other Britannica-maintained websites including merriam-webster.com. Merriam-Webster has served as America's leading dictionary publisher since 1831.

The lawsuit details three distinct ways Perplexity allegedly infringes copyrights. First, through data curation using software called "PerplexityBot" to crawl and scrape websites. Second, by copying copyrighted articles as inputs for its retrieval-augmented generation (RAG) model. Third, by generating outputs that contain verbatim reproductions, summaries, or paraphrases of copyrighted works.

Perplexity utilizes RAG technology, which connects large language models to external information sources to improve output quality. The process involves receiving user prompts, obtaining content from search indexes, combining prompts with retrieved content, and generating natural-language responses. According to Perplexity's founder and CEO Aravind Srinivas, this approach follows the principle that models should "only say what you can cite."

The complaint includes specific examples of alleged infringement. When users asked Perplexity about Merriam-Webster's definition of "plagiarize," the system reproduced the exact definition from the copyrighted Collegiate Dictionary. Similarly, requests for Britannica content on topics like Druids and quantum physics resulted in near-verbatim reproductions of copyrighted articles, including identical selections and ordering of information.

Investigators have documented Perplexity's use of undisclosed crawling methods that bypass website protections. According to a Cloudflare report from August 2025, Perplexity employs "stealth, undeclared crawlers" that impersonate Google Chrome browsers and utilize multiple IP addresses not listed in the company's official ranges. This allows the system to access content even when websites explicitly prohibit such access through robots.txt files and firewall rules.

The publishers emphasize the financial impact of this alleged infringement. Britannica's terms of use explicitly prohibit data mining and AI training without express written consent. The lawsuit states that Perplexity's outputs are "designed to eliminate the need for its users to visit the original content creators' websites," thereby reducing advertising and subscription revenues.

Perplexity has faced mounting criticism over its content practices. The New York Times and BBC have issued cease-and-desist notices, while Dow Jones filed a similar lawsuit in January 2025. Forbes described the company's approach as "cynical theft," and multiple publications have accused Perplexity of operating as a "plagiarism engine."

In response to criticism, Perplexity launched a Publishers' Program in July 2024 offering revenue sharing with content creators. The company has also secured licensing agreements with publishers including Time, Der Spiegel, and The Los Angeles Times. However, the publishers argue this program represents an attempt to "retroactively dictate the terms of a license" after already taking copyrighted material.

The lawsuit also addresses trademark violations, alleging that Perplexity falsely attributes AI-generated "hallucinations" to the publishers while displaying their trademarks. Additionally, the complaint claims Perplexity omits portions of articles without disclosure while suggesting complete reproduction.

Perplexity has maintained that its practices fall within fair use copyright protections. Chief Business Officer Dmitry Shevelenko stated the company operates "clearly within those bounds" of fair use and copyright law. However, the U.S. Copyright Office's May 2025 report on AI and copyright concluded that RAG systems are "less likely to be transformative" when generating summaries or abridged versions of copyrighted works.

The company's rapid growth underscores the financial stakes involved. Perplexity reached a $20 billion valuation in its latest funding round and reports serving over 100 million generative search results weekly. The startup, backed by investors including Jeff Bezos and Nvidia, has approximately 22 million active users across its platforms.## Why This Matters for the Marketing Community

This lawsuit represents a critical development for digital marketers and advertising professionals, as it challenges fundamental assumptions about how AI platforms can monetize copyrighted content. The legal action directly impacts how AI search engines can operate commercially, particularly as Perplexity has positioned itself as an alternative to traditional search platforms.

The case raises significant questions about the sustainability of AI-powered advertising models. Perplexity launched its advertising program in late 2024, featuring sponsored questions and side media placements while claiming to maintain unbiased search results. However, if courts find the underlying content sourcing practices illegal, the entire business model faces potential disruption.

The departure of advertising head Taz Patel in August 2025 after only nine months may reflect broader challenges facing the platform's monetization efforts amid mounting legal pressure. Marketing professionals who have invested in Perplexity's advertising ecosystem now confront uncertainty about platform stability and long-term viability.

The broader implications extend beyond Perplexity to the entire AI advertising landscape. As AI agents potentially replace humans as advertising targets, the legal framework governing content usage becomes increasingly critical. If publishers successfully defend their copyright claims, it could force fundamental changes in how AI platforms source and monetize information.

For performance marketers, this case highlights the importance of diversification across advertising platforms. The US Copyright Office's recent guidance on AI training suggests that licensing arrangements will become essential for sustainable AI operations, potentially increasing costs for platforms like Perplexity and ultimately affecting advertising pricing.

Timeline

• July 31, 2024: US Copyright Office releases Part 1 of AI Report addressing digital replicas
• October 2024: Dow Jones files major lawsuit against Perplexity AI over alleged RAG technology copyright infringement
• November 12, 2024: Perplexity announces first advertising test with sponsored questions and side media
• January 15, 2025: Google announces licensing partnership with Associated Press for AI training
• January 29, 2025: US Copyright Office releases Part 2 of AI Report examining AI-generated works
• May 1, 2025: US Copyright Office releases Part 3 of comprehensive AI report addressing AI training fair use
• June 4, 2025: Reddit files lawsuit against Anthropic for unauthorized Claude AI training
• July 9, 2025: Perplexity launches Comet browser for $200 Max subscribers
• July 13, 2025: Perplexity acquires OS.ai domain to build AI operating system
• August 25, 2025: US Attorneys General target AI companies for child safety failures
• August 29, 2025: Perplexity advertising head Taz Patel departs after nine months
• September 10, 2025: Encyclopædia Britannica and Merriam-Webster file lawsuit against Perplexity AI

Summary

Who: Encyclopædia Britannica, Inc. and Merriam-Webster, Inc. filed the lawsuit against Perplexity AI, Inc. in the Southern District of New York. The plaintiffs are represented by attorneys from Susman Godfrey L.L.P.

What: The publishers allege massive copyright infringement through three mechanisms: unauthorized crawling and scraping of websites, copying copyrighted content as inputs for RAG processing, and generating outputs that reproduce or summarize protected works. The lawsuit also claims trademark violations through false attribution of AI hallucinations and undisclosed content omissions.

When: The complaint was filed on September 10, 2025, covering alleged infringement dating back to Perplexity's launch in 2022. The case builds on mounting legal pressure following previous lawsuits from Dow Jones and cease-and-desist notices from The New York Times and BBC.

Where: The lawsuit was filed in the United States District Court for the Southern District of New York. The alleged infringement affects content creators globally but particularly impacts publishers with significant New York-based operations and readership.

Why: The publishers seek to protect their substantial investment in content creation and prevent what they characterize as the "massive illegal transfer of revenue" from content creators to AI platforms. They argue that Perplexity's practices threaten the economic incentives necessary for producing high-quality, trustworthy information.