New York Times warns Perplexity: Not our content
The New York Times doesn't want to show up on the AI search Perplexity. However, they say there is no exclusive information.
(Image: Peshkova/ Shutterstock.com)
"Perplexity and its business partners have illicitly enriched themselves by using the Times' expressive, carefully written, researched and edited journalism without a license," the New York Times (NYT) wrote in a letter to AI Search. The Wall Street Journal reports on this. The NYT is already suing OpenAI and Microsoft on similar grounds. There are also accusations against Perplexity from Forbes and the Condé Nast publishing house.
Videos by heise
The NYT does not want its content to appear in Perplexity's AI search. Accordingly, the newspaper has excluded numerous crawlers using a robots.txt file. However, Perplexity is fighting back. The company claims that it does not scrape content to train its own AI models. They would only index the websites and factual content. "No one has the sole copyright to facts," The Verge quotes a spokesperson from Perplexity. In a statement, the AI search also explains that publishers also report on the same topics and rely on each other.
AI searches and the question of copyright
Perplexity has already entered into partnerships with a number of publishers, including Fortune, Time and the Texas Tribune, as well as Spiegel in Germany and Automattic, provider of Wordpress. The deal means that as soon as information from these partners is used to answer a query in the search engine, the partner receives a share of the advertising revenue. It is not yet known how much revenue has been generated and how much the publishers receive.
Perplexity sees its search as a paradigm shift and therefore calls it an answer engine. Partners also get a year's free access to Perplexity's enterprise subscriptions and developer tools, including the API, as well as access to Scalepost.ai. This is a start-up designed to monitor just such partnerships. Scalepost allows you to see how often articles have been used for search engine responses.
The New York Times has also already filed a lawsuit against OpenAI and Microsoft. On the one hand, this concerns the display of content, specifically content that was behind a paywall and was displayed by ChatGPT virtually one-to-one. Secondly, it concerns the use of the data for training AI models – and the question of whether this was legal. ChatGPT also has an AI search function that uses its own crawlers. Website operators must explicitly exclude these if they do not want their content to be used for search results. So far, there is no remuneration for their use. Nevertheless, the links to the sources are displayed more prominently than was the case at the beginning of ChatGPT.
(emw)
