A new AI start-up has signed deals with several major global news publishers convinced by its plan to share revenue each time their content is used to generate an answer – before it has even launched.
The Financial Times, Fortune, Axel Springer and The Atlantic have all agreed to license their content to Prorata.ai.
The company says it has created technology that enables generative AI platforms to “accurately attribute and share revenues on a per-use basis with content owners”.
It “analyses AI output, measures the value of contributing content and calculates proportional compensation” meaning revenues can be apportioned to the rightful copyright holder even when multiple sources are used in an answer.
Related
Prorata is planning to launch its own chatbot in the autumn and will share half the revenue from subscriptions with its licensing partners.
Prorata says the technology will also be available to license to other generative AI companies like OpenAI and Anthropic.
Chief executive Bill Gross said: “Current AI answer engines rely on shoplifted, plagiarised content. This creates an environment where creators get nothing, and disinformation thrives.
“Prorata is pro-author, pro-artist and pro-consumer. Our technology allows creators to get credited and compensated while consumers get attributed, accurate answers. This solution will lead to a broader movement across the entire AI industry.”
Gross is credited for pioneering the keyword-based pay-per-click model of advertising in search engines, having created GoTo.com which was sold to Yahoo in 1998.
Gross now leads Idealab Studio, a technology incubator which has launched Prorata.
The new company’s senior leadership team includes chief strategy officer Annelies Jansen, formerly director, new media at The Telegraph and digital director at The Times during the digital revolution of the early 2000s and later VP, global commerce partnerships at Meta.
What publishers are saying about Prorata’s ‘fair attribution’ plan
The publishers on board so far have all signed other major deals with other AI platforms.
Axel Springer was one of the first publishers to sign with OpenAI, agreeing a partnership in December for its content to be summarised in ChatGPT with links and attribution and to be used in the training of the model. It later signed an expanded partnership with Microsoft with a plan to partner on new AI-driven chat experiences.
Deputy chief executive Jan Bayer said of Prorata that Axel Springer is encouraged by its “commitment to fair attribution and compensation for news publishers and content owners, and we look forward to collaborating with their team to develop and refine their platform.
“It’s imperative that innovative teams developing AI-driven technology do so in partnership with journalism organisations.”
The Financial Times became the first major UK-based newsbrand to sign a deal with OpenAI in April. Chief executive John Ridding said this week: “There is an urgent need and opportunity to align the incentives of AI platforms and publishers in the interests of quality journalism, the reader and respect for intellectual property.
“Prorata’s approach – identifying source material and sharing resulting revenues between technology companies that use it and publishers that create it – can help develop a healthier and fairer information ecosystem that encourages accurate and authoritative journalism and rightly rewards those who produce it.”
[Read more: FT CEO: News orgs ‘have leverage and should insist on payment’ from AI companies]
The Atlantic signed its own OpenAI deal in May, meaning its articles will be discoverable with links and attribution within ChatGPT. It later backed the AI company’s prototype search engine SearchGPT, which is in testing with publishers.
But the deal provoked some backlash from staff who feel, as the authors of the work being used, they should have a seat at the table in such negotiations. In a letter to The Atlantic’s chief executive Nicholas Thompson and editor-in-chief Jeffrey Goldberg on 1 August, they said: “Our editorial leaders say that The Atlantic is a magazine made by humans, for humans. We could not agree more.
“Although we understand that there may be a place at The Atlantic for artificial intelligence, Atlantic staffers must have a voice in how it affects our work.”
Days later, Thompson said in support of the company’s latest licensing deal: “Prorata is working to solve one of the most important issues in AI: how can LLMs properly credit and compensate the publishers of the work they depend on.
“With this partnership, we are working with them to establish first principles for media and AI – in a way that values, respects, and protects the exceptional talent and intense work of journalists and creators. We hope to see this approach to permissions, content controls, clear attribution, and fair value become the industry norm.”
Finally, Fortune was last week announced as one of a number of launch partners for AI search chatbot Perplexity’s revenue-sharing plan. That model means that when Perplexity introduces advertising via sponsored related questions in the next few months, signed-up publishers will get a share of the revenue generated by interactions where their content is referenced.
Chief executive Anastasia Nyrkovskaya said the team at Fortune believes Prorata “place a high priority on proper attribution and compensation for quality content.
“The goal is to work together to bring Fortune’s world-class business journalism, valuable archives, and authoritative lists to new audiences with Prorata’s innovative approach to AI.”
Prorata said it is also in “advanced discussion” with other global news publishers as well as media and entertainment companies and more than 100 authors.
Email pged@pressgazette.co.uk to point out mistakes, provide story tips or send in a letter for publication on our "Letters Page" blog
