Mo' Money, Mo' Problems: Perplexity Seeks Funds, Finds Trouble
Fundraising draws fire from publishers
If it’s true the that road to hell is paved with good intentions, then Perplexity AI has found the path.
The AI-powered search engine, backed by Jeff Bezos, has been a pioneer in retrieval augmented generation (RAG), as it angles to dethrone Google. It uses search technology to find sources of information on the web in response to users’ queries. Rather than simply presented a list of links to relevant sites as with a traditional search engine, however, RAG engines use information culled from those sites to generate their own responses to users’ questions, in some cases without links or attribution to the original sources, potentially depriving those websites of valuable search traffic (Perplexity includes links to sources below its summaries).
In a bid to play nice and avoid the friction many other AI companies have encountered with publishers, Perplexity promised earlier this year to begin sharing advertising revenue with publishers when their content is used to generate query responses, once it has such revenue to share (its current main source of revenue is a $20-a-month subscription for access to a more powerful version of its technology).
Good intentions have not produced good karma, however.
Last week, the New York Times sent Perplexity a “cease and desist” letter through its law firm demanding the AI company immediately stop using Times content to create its summaries. The letter also asked for information on how Perplexity was accessing the Times’ website despite the publishers’ technical measures to block unauthorized scraping.
The letter gave Perplexity a deadline of October 30 to respond. In an interview with the Wall Street Journal, Perplexity CEO Aravind Srinivas said the company plans to respond by the Times’ deadline.
“We are very much interested in working with every single publisher, including the New York Times,” he insisted. “We have no interest in being anyone’s antagonist here.”
Given the Times’ notably pugnacious legal history concerning the unauthorized use of its content by AI companies, however, the deadline is likely to be followed by litigation no matter how Perplexity responds.
The Journal itself, meanwhile, has beaten the Times to the punch. On Monday, its parent company, News Corp., which also publishes Barrons and the New York Post, filed a lawsuit against Perplexity in federal district court in New York leveling many of the same allegations as the Times’ letter, and seeking massive monetary damages from the AI company.
Perhaps coincidentally — though probably not — the legal moves by the publishers come as Perplexity is seeking to raise an additional $500 million at a valuation of $8 billion, more than double its current value. That’s a tiny fraction of the $6.6 billion recently raised by OpenAI, at a valuation of $157 billion. But the coincidence in timing between Perplexity’s fundraising and the Times' and News Corp.’s actions suggests an evolution in the publishers’ legal strategy, from a more or less exclusive focus on copyright infringement to framing a broader economic and finance-focused case against generative AI companies.
As the Times put it in its C&D letter, “Perplexity and its business partners have been unjustly enriched by using, without authorization, The Times’s expressive, carefully written and researched, and edited journalism without a license.”
Similarly, News Corp., in its complaint, alleges that Perplexity’s “core business model involves engaging in massive freeriding on Plaintiffs’ protected content to compete against Plaintiffs for the engagement of the same news consuming audience, and in turn to deprive Plaintiffs of critical revenue sources.”
It also claims Perplexity’s propensity to “hallucinate” facts while falsely attributing them to News Corp. publications harms those mastheads’ trademarks. It cites an instance where Perplexity referenced a Wall Street Journal article about the U.S. arming Ukraine-bound F-16 jets with advanced weaponry, saying Perplexity attributed quotes to the article that never appeared in it.
As for OpenAI’s far larger recent funding round, the Times is already in litigation against the company and News Corp. signed a $250 million licensing deal with the ChatGPT maker earlier this year, so that base is already covered. Some of that newly raised cash, inevitably, will also be used to pay for more licensing deals and settle some of the many lawsuits currently pending against OpenAI, including perhaps the Times’ suit.
The key question for AI companies going forward will be whether rights owners’ efforts to associate funding explicitly with liability could cool investors’ ardor for AI. The question for rights owners will be whether suppressing the appetite for AI risk among early-stage investors’ could unintentionally promote greater concentration within the AI industry, to where only the legacy Big Tech companies are left.