The most widely distributed newspaper in Japan, Reading and Selling News, has brought a copyright infringement suit against Perplexity, a creative, artificial and original company, for illegal seizure of its articles. The suit, which was filed with the Tokyo District Court on 7 August, marked the first time that a major news publisher in Japan had launched a copyright challenge against AI.

In its petition, Read and Sell News claimed that, based on an analysis of its corporate server log, Perplexity had visited 11,467 articles on its website between February and June of this year. In Read and Sell News, Perplexity alleged that it used online capture technology to respond to user queries by accessing, without authorization, articles that were copyrighted by newspapers.
In particular, it was stated that Perplexity had violated its “reproduction rights” and “the right to communicate to the public”. These two rights are the core rights granted to copyright owners under Japanese law to enable them to control the reproduction and dissemination of works. The action sought compensation for nearly 1.5 billion yen (approximately $15 million) and stopped copying its articles.

It is noteworthy that Japan ‘ s copyright law allows AI developers to use copyright-protected materials to train models without permission. This lax provision was directly derived from the amendment of the Copyright Act of Japan in 2018, which was designed to encourage the development of AI in the field of science and technology in Japan. However, the law does not permit the reproduction of these works as a whole or their distribution in a manner “unreasonablely prejudicial to the interests of the copyright owner”.
In his statement, a speaker from Perplexity stated: “We deeply regret the misunderstandings that have taken place in Japan. We are currently trying to understand the nature of these ideas. We attach great importance to this matter because of Perplexity’s commitment to ensuring that publishers and journalists can benefit from the new business model that emerged from the AI era.”

Last fall, two publishers under the US News Group, Wall Street Journal and The New York Times, took similar legal action against Perplexity. Outside the United States, however, Perplexity has so far faced less legal review. Competing revenue-producing AI companies, including OpenAI and Meta, have faced copyright infringement litigation from major international publishers.
In India, a joint copyright tort suit against OpenAI included numerous old Indian press publications such as The Indian Express, The Hindus and Recent India. In France, the association of its principal authors and publishers has brought proceedings against Meta for the existence of an economic “pathological act”.

In May of this year, the Press Association of Japan issued an open letter condemning AI’s “free-riding” use of its copyright-protected material and warning them to stop cyber-crawling. At that time, the Association wrote that the status quo “may cause considerable damage to the operations of the press agencies”. The public’s right to know may be impeded if the quality of news that underpins democracy is reduced.”

