The AI 'Doom Loop': How LLMs Are Cannibalizing Their Own Content Supply Chain

In the creator economy, monetization depends on a functional pipeline: creators produce value, platforms distribute it, and audiences engage. However, as 404 Media first reported, unsealed court filings from the *New York Times vs OpenAI* lawsuit reveal that the architects of generative AI may have built a system that systematically dismantles this pipeline.

According to reporting from 404 Media, internal documents and depositions from Microsoft and OpenAI executives describe a predatory relationship with the web's content. An internal Microsoft document explicitly warns of a "doom loop" created by AI content strategies. This loop occurs when AI products steal from human creators and subsequently cannibalize the clicks those creators rely on for revenue, effectively destroying the business models of the very entities providing the training data.

From a monetization standpoint, the risk is systemic. Microsoft executive Brent Hecht noted that LLMs extract content without providing a mechanism to distribute economic value back down the supply chain, which "necessarily threatens the economic stability" of content creators. The document further characterizes the LLM business as a product that "destroys its supply chain," noting the unusual nature of an end-product threatening its own essential suppliers.

The impact on traffic—the primary engine for creator monetization—is already evident. 404 Media reports that Microsoft CEO Satya Nadella testified under oath that clicks to news sites, including the *New York Times*, cratered by more than 90 percent on Bing after content was ingested. This suggests that even when links are provided, the utility of the LLM replaces the need for the user to visit the source. An OpenAI software engineer testified that regardless of how prominently links are displayed, users simply will not click.

Beyond the traffic collapse, the filings reveal a conscious effort to bypass protections. 404 Media reports that OpenAI developed a "hack" to circumvent the *New York Times* paywall, an action OpenAI cofounder Greg Brockman reportedly described as "ah, nice."

Ultimately, these admissions suggest that the current AI trajectory is not just a copyright dispute, but an existential threat to human labor. Jack Clark, OpenAI's policy director, wrote that the firm was building systems designed to replace the labor of the individuals who shape the culture of society. Microsoft's own policy documents admit that generative AI could significantly disrupt the employment of the people who generated the training data, while an internal Microsoft document described the process as "the largest theft of labor in human history."