AI and Media Titans Quietly Hash Out Future of Content Licensing

The world's leading AI tech companies have been engaging in hush-hush meetings with leading media moguls, the Financial Times revealed. Their goal? To hash out a landmark deal on the use of news content for training generative AI models such as Google’s Bard and OpenAI’s ChatGPT. 

In recent months, representatives from AI companies OpenAI, Google, and Microsoft have been in talks with news executives from notable publishers such as News Corp, The New York Times, Axel Springer, the Financial Times, and The Guardian as they grapple with the challenging topics of generative AI, copyright, and appropriate compensation.  The hope is to craft a milestone agreement that would have media companies receiving fees for providing continued access to their content used to train AI models.

These talks are unfolding under a rapidly changing landscape for content. In recent months, copyright infringement lawsuits against AI companies have emerged while legislative regulations such as the EU’s AI Act, which demand more transparency about the use of copyrighted content in AI model training. Companies like Reddit and StackOverflow, whose vast bodies of data have been used to train AI models, have also announced that access to their valuable datasets will now require payment.

The Financial Times, one of the media players involved in the dialogue with AI companies, emphasized, "Copyright is a crucial issue for all publishers. As a subscriptions business, we need to protect the value of our journalism and our business model. Engaging in constructive dialogue with the relevant companies, as we are, is the best way to achieve that."

Much of the early discussions have centered around a potential pricing model for news content access, with one option in the form of an 'unlimited use' model priced between $5M to $20M annually. Critics of this approach argue that this payment model may exclude smaller media organizations and regional news publications.

Also under consideration is also a quantitative model, somewhat akin to how streaming platforms like Spotify pay music royalties each time a track is played. But such a model poses complexities in calculating appropriate compensation.

Driving the conversation from the AI side is Google, which has engaged in extensive dialogues with news outlets and already proposed potential licensing options. An executive at a newspaper group explained the progress thus far: "[Google] has accepted the principle that there needs to be payment, but we have not got to the point of talking zeros."

OpenAI's CEO Sam Altman has also expressed his support for compensating content owners. “We think that content creators, content owners need to benefit from this technology,” said Altman in his appearance before the US Senate last month. “Exactly what the economic model is, we’re still talking to artists and content owners about what they want. I think there’s a lot of ways this can happen. But very clearly, no matter what the law is, the right thing to do is to make sure people get significant upside benefit from this new technology. And we believe that it’s really going to deliver that, but that content owners, likenesses, people totally deserve control over how that is used and to benefit from it.”

Despite the nascent progress, media executives are confident of reaching a viable deal. Many are determined not to be caught off guard as they were in the early years of the internet era, when news organizations quickly found revenues shrink as consumption moved online. 

As Mathias Döpfner, Axel Springer's chief executive, puts it, AI companies "know that regulation is coming, and they are fearful of it." He added, “It is in the interest of all parties to come up with a solution for a healthy ecosystem. If there is no incentive to create intellectual property, there is nothing to crawl. And artificial intelligence will become artificial stupidity.”