Microsoft Exec Called AI Scraping 'The Largest Theft of Labor in History'

Microsoft called AI scraping 'the largest theft of labor in human history' in newly unredacted court filings, exposing a glaring contradiction: the company warned internally about the damage to publishers while its partner OpenAI built datasets from paywalled content. The revelation, reported by TechCrunch, pulls back the curtain on how Big Tech really views data ethics.

Context: The quiet war over training data

The unsealed documents stem from The New York Times lawsuit against OpenAI and Microsoft for using millions of copyrighted articles to train models like GPT. For months, key passages were blacked out. Now, the full filings show Microsoft executives and lawyers openly discussing the devastating impact on publishers—even as they continued scraping paywalled sites like The Times.

Details: Internal emails that say it all

The unredacted filings include emails where a Microsoft executive labels OpenAI's data practices as 'theft'. According to TechCrunch, both companies scraped paywalled Times content, built datasets from it, and warned internally it would gut publishers. The irony is sharp: Microsoft, which also owns MSN and licenses content, knew exactly what was at stake while profiting from the same system.

Analysis: Cynicism as a business model

What these documents reveal isn't just a legal spat—it's the structural hypocrisy of the AI industry. Big Tech knows massive uncompensated scraping is unsustainable, yet it does it until someone sues. The winners are AI companies getting free data; the losers are creators and media outlets whose work fuels products that then compete against them. My take: if Microsoft called it 'theft' in private, there's no defense in public.

Implications: Toward a new data social contract

This case could force regulatory change: mandatory licensing, compensation for publishers, or even the collapse of business models built on free scraping. If courts rule against Microsoft and OpenAI, the precedent will force the entire industry to pay for data. Publishers cheer, but the cost will trickle down to users. The battle for ethical AI is just beginning.

If companies know it's theft and do it anyway, should the law treat them as repeat offenders?

Source: TechCrunch