In a major legal case, **Microsoft** and **OpenAI**, a company in which **Microsoft** holds a 27 percent stake, are arguing before the Southern District Court of New York that using copyrighted works to train large language models is a "transformative use." This argument could set a precedent for how copyright law applies to artificial intelligence, potentially creating a "fair use" exception for training AI systems. The case involves several prominent authors and the New York Times, who claim that **Microsoft** and **OpenAI** used their works without permission or payment to develop competing products. At the heart of the dispute is whether training AI models using copyrighted texts qualifies as fair use under copyright law. **OpenAI** has referenced a 2025 court ruling by Judge William Alsup, who determined that using protected books to train AI models can be considered fair use. Alsup argued that such use does not harm the commercial value of the original works and leads to the creation of new content that benefits the public. **OpenAI** emphasized that their goal is to build statistical language models that generate new texts, not to copy or plagiarize existing works. On the other side, the authors and the New York Times argue that using their works without permission harms the market for original content and diminishes the incentive for human creativity. Judge Vince Chhabria, who has also been involved in the case, acknowledged the transformative nature of the AI training but pointed out that such practices remain illegal without proper authorization or compensation. The case has grown beyond just copyright law, touching on broader issues of economic competition, with authors claiming that AI models like those developed by **Microsoft** and **OpenAI** directly compete with original publications. **Microsoft** has presented evidence from a study analyzing conversations with its AI assistant, **Pilot**, which found no significant use of the 212 books central to the case. The legal proceedings have been consolidated in the New York court to ensure consistent rulings, reflecting the complexity and importance of the issues being debated. This case could shape the future of AI development and how copyright law adapts to emerging technologies.