A Study on News Works used for Generative AI learning

The development of Generative AI technology raises an important issue in the copyright field. Recently, in the U.S., Europe, and Korea, copyright infringement lawsuits have been filed against Generative AI companies such as Open AI, Meta etc., because of using news works as deep learning data without permission. In the U.S., the NY Times filed a copyright infringement lawsuit against Open AI, and eight major newspapers also filed a lawsuit against Open AI and MS, saying that AI Chatbot often displayed news published exclusively for subscribers for ChatGPT and did not indicate source links. I n K orea, three broadcasters f iled a l awsuit a gainst t he company Naver, claiming that they used news works to learn Generative AI without consent, therefore they need consent, and must pay fair compensation. Tech companies are arguing that the use of learning data belongs to fair use, so it is not copyright infringement, but should be fair use, because Generative AI technology brings lots of benefits to the society, and learning data, including news works, must be used for AI learning for the public interest. Owing to the development of Generative AI technology it is a time when we are in a dilemma of promoting Generative AI industry and protecting the copyright of copyright holders. Therefore, it is important to analyze the comparative legal cases, and we look forward to providing legislative alternatives following as discussions on this issue will actively take place.