OpenAI and Microsoft knew they were starting a ‘doom loop’ for the web

Unsealed court documents from the New York Times lawsuit reveal that internal OpenAI and Microsoft communications acknowledged their data scraping practices could initiate a damaging cycle for the internet. These records describe the large-scale collection of information for training models as a potential threat to the web's stability. This evidence is now part of ongoing litigation concerning whether the companies' use of copyrighted content constitutes unauthorized labor appropriation.