Canada’s leading media organizations have filed a class-action lawsuit against OpenAI, which develops artificial intelligence technology. The lawsuit alleges that the company extensively used journalistic content without permission to train large language models. It is emphasized that OpenAI uses these contents during the training of models such as ChatGPT, but does not pay any permission or copyright fee.
Case; It was covered by the country’s leading media outlets, including The Globe and Mail, Toronto Star, National Post, CBC/Radio-Canada and The Canadian Press. These organizations argue that such uses by OpenAI clearly violate copyright laws. The plaintiffs are seeking damages of C$20,000 for each article. It is stated that if these demands are accepted, the total compensation amount could reach billions of dollars.
The ongoing debate around copyrights and artificial intelligence
Media organizations oppose OpenAI’s statements that it uses public data, arguing that such content cannot be considered within the scope of “fair use”. Postmedia, which owns the National Post, said in a statement: “Journalism is in the public interest. However, it is illegal to use these contents without permission for commercial gain,” he said.
OpenAI had previously argued in similar cases that its contents were collected from publicly available sources and that this was based on fair use principles. Company spokesman Jason Deutrom stated that they cooperate with journalistic organizations and said, “We offer various facilities to publishers in displaying, citing and linking their content on ChatGPT.” However, these statements are found inadequate by many organizations in the media sector.
Meanwhile, it is known that some media organizations have taken a different approach and made licensing agreements with OpenAI. For example, organizations such as the Associated Press, Financial Times and Time have formally agreed to regulate the use of their content in artificial intelligence models. In contrast, plaintiffs in Canada reject such cooperation and prefer to pursue their rights through legal means.
The case in Canada is considered part of a growing legal battle by media companies around the world against artificial intelligence companies. For example, in the USA, many publications such as The New York Times, The Intercept and Chicago Tribune filed a lawsuit against OpenAI with similar accusations. However, some cases were rejected by the courts due to lack of sufficient evidence. Despite this, it is stated that plaintiffs can revise their complaints and re-apply.
In addition to all this, the media industry’s reactions to big language models are not limited to legal steps. Some media companies also express concerns about how artificial intelligence technologies could transform content production processes and the media economy. In particular, it is frequently emphasized that the unauthorized use of original content may weaken journalistic values and lead to loss of income in the sector.
This case filed in Canada sets an important example at a time when the development of artificial intelligence technologies and copyright laws conflict. The legal status of the contents used for training large language models and whether compensation will be paid to the owners of these contents may come to the fore more frequently in the future.
It is thought that such legal processes between artificial intelligence companies and media organizations will shape cross-sector relations in the future. In particular, it seems that issues such as the limits of the fair use principle and the balance between public interest and commercial gain will be discussed further. The case in Canada is a remarkable example of how far media organizations can go in their fight to protect content rights.