Unsealed court documents in the New York Times’ lawsuit against OpenAI and Microsoft reveal that both companies internally recognized the destructive potential of their data scraping practices, characterizing them as a “doom loop” for the web. These admissions, first reported by The Verge, show the companies knew about the controversy surrounding AI model training.
Microsoft’s Director of Applied Science, Brent Hect, went further in the documents, allegedly calling the use of news content “an astonishing theft of unprecedented proportions” and possibly “the largest theft of labor in human history.” This specific phrasing, detailed by The Hindu, supports the New York Times’ claims that AI developers knew their actions were problematic. The Times initiated its legal action against OpenAI and Microsoft in August 2026, alleging widespread copyright infringement. The new internal communications suggest the companies understood the risks and ethical questions of training large language models on vast internet data, including copyrighted journalistic content.
The “doom loop” described in the internal documentation refers to a scenario where AI models, trained on existing web content, generate new content that then floods the internet. This AI-generated material could dilute the value of original human-created content, making it harder for creators to be discovered or compensated. In turn, this could disincentivize the creation of high-quality original content, leading to a degraded information environment and a self-reinforcing cycle of diminishing returns for both content creators and AI models. The internal warnings suggest OpenAI and Microsoft understood this potential degradation even as they pursued their development strategies.
Public Opinion Sours on AI and Its Infrastructure
These internal concerns about content ethics and the web’s future arrive as public sentiment towards AI development shows increasing skepticism. A recent poll conducted by The New York Times and Siena University found significant opposition to the infrastructure required to power AI. According to The Verge, the poll, released on Tuesday, surveyed 1,503 likely voters. Sixty-one percent of those surveyed opposed new data centers designed to power AI technologies.
This finding aligns with what politicians are observing and responding to. It indicates broader public uneasiness beyond intellectual property disputes. The poll suggests that the tangible elements of AI infrastructure, such as data centers, are becoming points of local and national contention. Public opposition creates a challenging environment for AI companies. They face legal challenges over content acquisition and public resistance to their physical footprint.
The New York Times has not only launched a significant lawsuit against major AI players but has also reported on growing public discontent with the technology at the heart of that suit. This dual role shows a significant tension in AI development. While AI companies push for rapid innovation and deployment, their methods and infrastructure face increasing scrutiny from both legal systems and the general populace. The contrast between internal corporate warnings of a “doom loop” and widespread public opposition to data centers shows a shared concern about the unbridled expansion of AI.
The unsealed court documents, with their candid internal admissions, could significantly influence the ongoing litigation between the New York Times, OpenAI, and Microsoft. Brent Hect’s blunt assessment of “the largest theft of labor in human history” provides direct evidence that the companies were aware of the scale and nature of their data acquisition, potentially complicating their defense. As the case progresses, these internal records will likely serve as crucial evidence about the early decisions made by AI developers.
What did unsealed documents reveal about OpenAI and Microsoft?
Unsealed court documents in the New York Times’ lawsuit against OpenAI and Microsoft show the companies internally warned that their data scraping practices could create a “doom loop” for the web and characterized the actions as “the largest theft of labor in human history.”
Who made specific claims about “theft of labor”?
Microsoft’s Director of Applied Science, Brent Hect, allegedly called the scraping of news content “an astonishing theft of unprecedented proportions” and possibly “the largest theft of labor in human history.”
What did a recent poll show about public sentiment on AI infrastructure?
A poll by the New York Times and Siena University found that 61 percent of 1,503 likely voters surveyed oppose the construction of data centers to power AI technology.
Compiled by Launch91 Desk from the sources linked above. More about Launch91.
