Newspapers Vs. ChatGPT— "If you can't beat them, join them"
Newspapers Seek a Finger in the ChatGPT Pie Even as AI Feasts on Free News Content.
The Growing Tension Between News Publishers and Generative AI
The world of artificial intelligence is undergoing an unprecedented gold rush, creating a climate of uncertainty and raising pertinent questions about the future of online information. On the one side are tech firms and their powerful generative AI models; on the other, source publishers and newspapers demanding equitable compensation. As Bilawal Sidhu, our honorary Tech Adviser based in Austin, Texas, aptly put it: “When cutting-edge technology disrupts established and complacent industries, it's typical for them to cry foul before eventually seeking a sliver of the new pie, even if it is wafer-thin.”
Newspapers Demand a Slice of the AI Market
A handful of major newspapers, including the likes of the New York Times, Reuters, and The Washington Post, have entered into talks with OpenAI, the creator of ChatGPT, over the use of their digital news stories. These publications are no longer content (pun intended) to let their meticulously researched and verified articles serve as free fodder for training AI models without seeing any financial benefits. Bloomberg Intelligence estimates the market for generative AI to reach a staggering $1.3 trillion by 2032, prompting these publishers to seek their share of the revenue pie.
The Countermeasures: Blockers and Licensing Agreements
Since August, over 500 news organisations have installed blockers that prevent their content from being collected and used by ChatGPT for training. Current discussions focus on compensating these publishers through direct payment and the potential increase in web traffic by displaying links to individual news stories in ChatGPT responses. An OpenAI spokesperson confirmed that the company is in talks with the newspapers but clarified, “None of the company’s practices have violated copyright law. Any deal would be for future access to content that is otherwise inaccessible or display uses that go beyond fair use.”
The Paradox of Content Accessibility
The situation presents a paradox for the news agencies. If the publishers make login credentials mandatory to access their content, they risk losing existing as well as potential readers who may not discover their articles in the first place. On the other hand, freely accessible content runs the risk of being used by generative AI models without any financial compensation either to the content-creators or to the publishers.
The Shifting Power Dynamics
Other platforms like Reddit are also grappling with the same dilemma. Few believe that it can survive without search engine visibility, but the question remains whether such a stance is sustainable in the long run, or whether this should preclude it from demanding a fair compensation. Moreover, leading AI firms are already battling copyright lawsuits from authors, artists, and software developers, since the existing laws and regulations are inadequate or simply unenforceable globally, in a consistent manner.
Conclusion: The Road Ahead
As the tension between AI companies and publishers intensifies, both sides are moving cautiously. OpenAI’s willingness to negotiate could be seen as an attempt to strike deals before the legal framework around this issue crystallises. Publishers, on the other hand, are wary but optimistic about leveraging their collective strength.
As generative AI continues to evolve, the conversations around data rights and fair compensation are only going to get louder. What remains to be seen is how these erstwhile ‘free’ resources navigate this complex terrain to establish an equitable relationship with the burgeoning AI industry.
