Generative AI Proposals: Use High Ethical Standards in Domestic Development

Editorial

14:00 JST, December 16, 2025

Generative artificial intelligence, which is evolving rapidly, now impacts all aspects of people’s lives, extending far beyond the business community which seeks to enhance work efficiency.

Meanwhile, cases of AI being abused for crimes such as fraud and spreading fake information continue unabated.

To build a society in which humans and AI can coexist, the need for high ethical standards and moral values cannot be taken lightly. For this reason, it is essential for Japan itself to take the lead in AI development and in promoting its diffusion.

The Yomiuri Shimbun has compiled a set of proposals about generative AI. It urges the government to take an active lead in such fields as the development of “large language models,” which recognize the meaning and usage of words to generate text, and managing the data that AI learns from, thereby enhancing Japan’s autonomy.

In a survey of private companies conducted by the Internal Affairs and Communications Ministry this year, 55% of respondents answered they “use generative AI in their business operations.” However, the United States and China lead in AI development. The AI used in Japan is almost exclusively developed overseas.

Over-reliance on foreign-made AI is undesirable in terms of economic security. Concerns have been raised that access to AI could be restricted if Japan were to clash with the country in which the AI in question is developed.

In China, AI developers are prohibited from disseminating content that runs against the policies of the administration of Chinese President Xi Jinping. There was a case in which China-made AI was asked, “Are the Senkaku Islands Japanese territory?” and it gave this answer: “They are an inherent part of Chinese territory.”

If AI learns extreme concepts or assertions that contradict facts, it risks misleading people or threatening societal standards. Development of trustworthy, domestically produced AI based on sound learning data is an urgent priority.

Japan is a tolerant nation that does not discriminate against people on factors such as race and where differing perspectives are respected. The ethical standards that have taken root in Japanese society should be reflected in AI development.

The proposals also highlight that it is indispensable to establish operational rules and laws concerning AI.

Who bears responsibility among the driver, the automaker and the AI developer if an autonomous vehicle using AI causes an accident? A similar problem would arise if a doctor uses AI for diagnosis and makes an erroneous judgment.

Leaving various issues unaddressed and pursuing only the convenience of AI while promoting its diffusion will leave seeds of future trouble behind.

Moreover, when opportunities to use AI increase, there is a risk of becoming overly reliant on its answers, leading to a decline in critical thinking and expressive abilities. It is crucial to learn about these negative aspects from childhood onward.

(From The Yomiuri Shimbun, Dec. 16, 2025)