Microsoft Docs Reveal AI Ethics Concerns: 'Existential Threat' to Publishers

Internal Microsoft documents have revealed concerns about the ethics of web scraping and AI training in generative AI technology. A June 2023 memo from ChatGPT development head Nick Turley described AI as 'an existential threat' to publishers, stating that it becomes 'more and more substitutive' as it improves.

Turley also wrote in a February memo that year that AI products are 'largely substitutive, period,' which goes against the fair use doctrine of transforming but not replacing original work. Microsoft employees debated whether OpenAI was committing 'the largest theft of labor in human history,' and internal memos warned that large AI models 'destroy their supply chain.'

The New York Times has obtained these documents as part of its lawsuit against Microsoft and OpenAI, which argues that the companies' use of web scraping and AI training is a form of labor theft. While Microsoft claims that Hecht was hired to present diverse perspectives, it's clear that he and other employees were concerned about the impact of their technology on the publishing industry.