The New York Times is suing Perplexity for copyright infringement

The long-simmering battle between legacy media and AI has officially moved from training datasets to live search engines.

The New York Times filed a copyright infringement lawsuit against AI search startup Perplexity. It marks the publisher's second legal offensive against an AI firm, following its high-profile suit against OpenAI and Microsoft. The move came in the same week the Chicago Tribune also filed its own suit against Perplexity.

Bypassing paywalls and pinning hallucinations

The escalation didn't happen overnight. Just over a year ago, The Times sent a cease-and-desist letter to Perplexity, and the two sides have traded warnings over the past 18 months without reaching a deal.

In the filing, The Times claims Perplexity creates commercial products that directly substitute for original journalism without permission or compensation. The core grievance focuses on Perplexity's retrieval-augmented generation engine powering its chatbots and Comet browser assistant.

According to Times spokesperson Graham James, RAG allows Perplexity to crawl the web, bypass paywalls in real time, and deliver premium subscriber-only reporting directly to non-paying users. The lawsuit adds that Perplexity often generates near-verbatim reproductions or abridgments, and even hallucinates false statements while attributing them to The Times, eroding its brand reputation.

Perplexity fired back dismissively. Head of communications Jesse Dwyer told TechCrunch that publishers have sued every new technology for a hundred years, from radio to social media. Fortunately it never works, Dwyer argued, or we would all still be communicating via telegraph.

Mounting legal pileup

Perplexity's legal inbox is getting crowded. News Corp sued the company last year over Wall Street Journal and New York Post content. In 2025 alone, Encyclopedia Britannica, Merriam-Webster, Nikkei, Asahi Shimbun, and Reddit have all launched complaints or legal challenges.

Outlets like Wired and Forbes previously accused the startup of scraping sites that explicitly prohibited scrapers, a claim internet infrastructure provider Cloudflare later verified.

Perplexity has tried to blunt the backlash with revenue-sharing schemes. Its Publishers' Program promised advertising cuts to partners like TIME and Fortune. In August, it rolled out Comet Plus, pledging 80 percent of its five-dollar monthly fee to publishers, alongside a licensing pact with Getty Images.

Yet top-tier publishers clearly aren't sold on micro-dividends. The Times isn't anti-AI on principle; it signed a lucrative multi-year training licensing deal with Amazon earlier this year. The lawsuit is an aggressive negotiating tactic aimed at extracting formal, guaranteed licensing revenue.

Why it matters

For anyone relying on Perplexity as their primary research engine, this lawsuit threatens the core product experience. Perplexity's secret sauce has always been surfacing clean, real-time summaries of high-quality journalism. If court injunctions force it to blacklist premium paywalled publications, answers will inevitably get shallower, and the era of free-riding on investigative reporting will close.

For the AI sector, this blows up the assumption that RAG is a safe legal harbor. AI vendors long argued that pre-training on public data was fair use, treating runtime search summaries as an even safer workflow. The Times is directly challenging that logic by arguing that summarizing behind a paywall in real time is clear market substitution.

We think Perplexity's smug historical comparison ignores current legal realities. Just look at Anthropic, which recently agreed to a staggering 1.5 billion dollar settlement over copyrighted books. Ad-share experiments and 4-dollar revenue splits won't cut it for investigative newsrooms whose survival relies on hard subscriptions. Perplexity needs real licensing budgets, because laughing off lawsuits with telegraph jokes is an expensive strategy that rarely survives discovery.