Japan watchdog to probe AI search services’ use of news articles

Japan’s Fair Trade Commission (JFTC), the country’s antitrust watchdog, said that it will investigate the use of news articles by AI-powered online search services over antitrust / unfair competition concerns.

The JFTC suspects some companies of being in violation of the Antimonopoly Law, with the unauthorized use of news articles potentially constituting an abuse of a superior bargaining position, which is explicitly prohibited under the law.

In simple terms, regulators argue that the use of AI by search engines can put others at a disadvantage and hurt publishers’ business models.

“It has been pointed out that companies that have search engines with generative AI are using news articles without permission from media organizations and generating answers,” said JFTC Secretary General Hiroo Iwanari, according to The Japan News. “It has become necessary to examine how it affects competition.”

Reportedly, the investigation will cover services offered by major tech giants, including Google and Microsoft, ChatGPT developer OpenAI, US startup Perplexity AI, and Japan’s LY Corp, formed by the merger of LINE Corp and Yahoo! Japan.

Currently, AI can summarize search results based on a user query and generate an answer. This means that online users no longer have to click on news providers’ websites, which could potentially reduce their revenues.

In May, JFTC revealed draft guidelines for the Mobile Software Competition Act (MSCA), which is supposed to oversee smartphone software services of companies such as Google and Apple to help ensure fair competition for smaller companies.

Concerns about unauthorized use of scraped information from websites have already been circulating since the introduction of AI chatbots and AI summaries.

Website owners in various sectors, from news sites to global banks and universities, are increasingly blocking crawlers from accessing their content to try and hinder their ability to reuse it without permission.

Companies have also commonly filed lawsuits in cases of scraping. For instance, Reddit sued Perplexity and three other companies for allegedly scraping Google search results in which Reddit content appeared and stealing it to power its AI-based “answer engine” system. In turn, OpenAI was sued for scraping news content to train ChatGPT without permission by several Canadian news companies in 2024.