That’s not our language, either: “Doom loop” comes straight from the mouth of an unnamed Microsoft executive describing the situation, as quoted in an unsealed request for summary judgment that the Times‘ lawyers filed earlier today. (Circulated online by Digital Content Next’s Jason Kint.) The fuller quote isn’t any less bleak: “Our AI content strategy has started a ‘doom loop’ that will hurt the performance of our models and the entire web at the same time: It is highly unusual that an end-product threatens the economic foundations of its essential suppliers, but that is the situation we have created for our LLM business with respect to its ‘content supply chain.’”
This isn’t the only document Times attorneys quoted in their filing that made it clear that OpenAI and Microsoft knew, pretty shortly into this grand project, that they were probably going to be massively damaging the ability of organizations like the Times to produce the actual information and reporting that LLMs so gleefully serve up: Elsewhere, the brief quotes defendants’ own numbers on Bing click-through rates on New York Times articles, which have dropped 83-93% since the widescale adoption of its AI news summaries, while another quote lays out the self-defeating logic even more bluntly: “LLMs are a product that destroys its supply chain.”
That’s before even getting into the actual copyright issues that make up the bulk of the case, which is still ongoing, and which sees the Times‘ lawyers argue that none of this copying can constitute “fair use,” since OpenAI and Microsoft’s work clearly aims to substitute itself, to massive commercial gain, for the New York Times articles it’s copying. Shocking nobody, the Trump administration’s Department Of Justice recently waded into the quagmire on the tech billionaires’ side, arguing that “The United States has a strong interest in this court rejecting any argument that training LLMs on copyrighted texts violates copyright law.”
