Microsoft's AI Push Creates Catastrophic Doom Loop

The unsealed court documents from the New York Times' lawsuit against OpenAI have revealed that Microsoft knew about the damage their AI push would cause, but did it anyway. In December 2023, the New York Times sued OpenAI for using its copyrighted material to train their AI models, with Microsoft being dragged in as one of OpenAI's main backers.

Microsoft Director of Applied Science Brent Hecht had repeatedly warned that their strategy of scraping new stories to train AI was tantamount to 'an astonishing theft of unprecedented proportions.' He even said it could be the 'largest theft of labor in human history.'

Hecht contradicted Microsoft and OpenAI's stance that using copyrighted materials to train AI falls under 'fair use,' saying that it made 'a complete mockery of the idea of fair use.'

Microsoft CEO Satya Nadella admitted that their AI models are stealing from media companies, providing nothing in return, and creating a 'doom loop' that will destroy the web. Hecht agreed, saying that 'our AI content strategy has started a 'doom loop' that will hurt the performance of our models and the entire web at the same time.'