The U.S. Copyright Office is studying the use of copyrighted materials in AI training, suggesting that regulatory steps may be needed. The cost of developing AI software is already high, and requiring AI companies to pay market prices for all data they scrape online could potentially push many of them into financial difficulty. The lawsuit and the potential regulatory changes could have significant implications for the AI industry.
Key takeaways:
- OpenAI and Microsoft were sued by The New York Times for training their AI models on copyright-protected and paywalled content without compensation or disclosure.
- Some publishers, including the Associated Press and Axel Springer, have already reached commercial agreements to license their content to OpenAI, with deals ranging between $1 million and $5 million.
- OpenAI's AI model performance has reportedly declined due to the inability to use NYT-based language datasets for training, leading to complaints about its flagship ChatGPT products.
- The U.S. Copyright Office has initiated a study into the use of copyrighted materials in AI training, suggesting that legislative or regulatory steps may be required to address this issue in the near future.