The tv community CNN is taking purpose on the synthetic intelligence search engine Perplexity in a lawsuit over copyright infringement. As reported by the network’s Brian Stetler, the swimsuit, filed Thursday in a New York District Court docket, accuses the AI firm of copying and distributing CNN’s content material, together with over 17,000 of CNN’s tales, movies, photographs and different printed works.
Although that is CNN’s first authorized case towards an AI firm, the community joins different publishers who’ve sued the San Francisco-based startup, together with the New York Instances and Information Corp. Based on the swimsuit, CNN tried to strike a licensing take care of Perplexity, however these talks did not end in an settlement. CNN beforehand made a content material licensing take care of Meta final yr, the place the tech large compensates the media firm for utilizing its reporting and content material to answer queries on Meta AI.
AI merchandise usually scrape information publications and web sites to reply person questions with real-time knowledge, accelerating the collapse in site visitors and income to authentic sources.
In response to the lawsuit, Jesse Dwyer, Perplexity’s chief communications officer, instructed Stetler and different media retailers in an announcement: “You’ll be able to’t copyright info.” The US authorities’s Copyright Office states: “Copyright doesn’t shield info, concepts, techniques, or strategies of operation, though it could shield the best way this stuff are expressed.”
CNN mentioned in its personal assertion that an organization valued at tens of billions of {dollars} should not “steal from entities that create the unique content material Perplexity exploits” and that “industrial operators can and should pay to utilize it.”
A Perplexity consultant did not instantly reply to a request for remark.
AI copyright fits
Perplexity is one among a number of corporations, together with OpenAI and Anthropic, which were battling information publishers and media giants over copyright claims.
(Disclosure: Ziff Davis, CNET’s father or mother firm, in 2025 filed a lawsuit towards OpenAI, alleging it infringed Ziff Davis copyrights in coaching and working its AI techniques.)
Greater than 100 such lawsuits have been filed. However completely different conclusions have been reached as as to if coaching AI fashions on copyrighted knowledge counts as truthful use, mentioned Michael Goodyear, an associate professor at New York Law School. Concerns embrace how the coaching happens, what AI outputs include and whether or not there’s any aggressive hurt to copyright holders.
“No appellate courts have but weighed in on the viability of those copyright infringement claims towards AI corporations,” Goodyear mentioned.
Within the CNN case, he mentioned that Perplexity is right that info aren’t protected by copyright, however the best way CNN presents info may very well be.
“Even brief information articles would sometimes qualify for copyright safety below the low bar of required originality,” Goodyear mentioned. “The query turns into whether or not the 1000’s of instances of infringement CNN describes are copying complete paragraphs verbatim, or whether or not they’re paraphrasing or merely copying unprotectable info.”
AI licensing offers
As plunging web site site visitors has drained billions in writer income and triggered widespread media layoffs, AI corporations are aggravating the disaster. Based on a new report from the think tank Open Markets Institute, over the previous six months, the speed of AI crawlers bypassing paywalls and blocks has practically quadrupled, spiking from 3.3% to 12.9%.
That is partly why quite a lot of publishers signed AI content material licensing offers with tech corporations to monetize content material used to coach AI techniques. A method out for Perplexity could also be to renegotiate a licensing take care of CNN. Even when Perplexity has legitimate authorized arguments, a licensing settlement might shift from unauthorized scraping towards a formalized content material partnership.
Nonetheless, the Open Markets Institute report says that relating to AI content material licensing, information and content material creators are trapped in a double bind. The identical tech giants whose AI instruments are ravenous web sites of human site visitors at the moment are those gatekeeping the licensing offers meant to exchange that misplaced advert income.
