Microsoft exec called AI scraping 'the largest theft of labor in human history'
summary
Unredacted court filings from a copyright lawsuit reveal that executives at Microsoft and OpenAI privately admitted that scraping news content for AI training was equivalent to theft. These documents suggest the companies knowingly bypassed paywalls and stripped copyright notices to build their models. Such actions potentially undermine their legal defense of fair use by showing the AI products directly compete with and economically harm the original publishers. The internal admissions indicate a conscious effort to prioritize model growth over the legal rights of content creators.
key points
- Internal communications describe AI training as the largest theft of labor in human history and an existential threat to the journalism industry.
- Microsoft data shows a dramatic drop in click-through rates for publishers, creating a doom loop where AI substitutes the need for original sources.
- The companies allegedly used technical hacks to circumvent paywalls and deliberately removed copyright notices to prevent AI models from outputting them.
who should read it
Legal professionals and AI developers who need to understand the evolving landscape of copyright law and fair use. It is also highly relevant for journalists and publishers fighting for the economic survival of their content.
discussion skeptical
consensus
Commenters generally agree that existing laws often fail to account for the massive scale of AI training, which differs fundamentally from individual human learning. There is a strong sentiment that AI companies are exploiting legal loopholes to avoid paying creators, with many advocating for a licensing marketplace or royalty system. While some argue that AI benefits society as a whole, the majority believe that protecting intellectual property at scale is necessary to prevent the erasure of creative livelihoods.
top insight
The comparison between a police officer watching a street corner and a panopticon surveillance system effectively illustrates the 'scale' argument. It argues that while a single instance of an action may be legal or benign, automating that action at an infinite scale transforms it into a systemic violation of rights. This perspective shifts the debate from the legality of the act to the impact of its scale.
Generated by an LLM from the article and thread. Check the sources before relying on it.