LDP’s Proposal for AI Policy: Put Brakes on Excessive Focus on Development

Editorial

16:22 JST, May 11, 2026

It is not desirable to fall behind in the race to develop generative artificial intelligence. There is no denying that such impatience has led the government to underestimate the risks of AI.

The government should strengthen regulations by facing up to the negative aspects of AI.

The Liberal Democratic Party has compiled a draft proposal on AI policy. Regarding the AI Law that came into full effect last year, the party has called on the government to amend the law to ensure effective measures are taken, including penalties, if AI businesses fail to comply with government guidance and other regulations. 

The law stipulates government policies such as those that provide national support for AI development. It allows the government to investigate and provide guidance or advice to businesses in cases where AI-generated content results in copyright infringement or other problems. However, there are no penalties under the law even if businesses refuse to comply.

So far, the government and the LDP have promoted AI, regarding it as a pillar of growth. Nevertheless, the LDP has now called for strict measures against AI operators because some were reportedly uncooperative during the party’s hearing of opinions.

It is only natural to slap AI operators with penalties if they do not comply with government guidance and other rules. The AI Act established by the European Union in 2024 imposes severe penalties on businesses that fail to comply with the EU’s requests for information, among other matters.

There is no end to cases in which third parties have created characters that strikingly resemble manga protagonists and other characters by using AI companies’ video creation services and have posted unauthorized videos.

Currently, it is not easy to hold AI operators accountable for such incidents. This is attributable to a revision to the Copyright Law in 2018, which has allowed AI models to be trained on copyrighted works without permission from the copyright holders.

The government must not only rectify the shortcomings of the AI Law but also consider amending the Copyright Law once again.

Concerns regarding the development and use of AI go beyond these issues. The government has submitted to the Diet a bill to amend the Law on the Protection of Personal Information, a revision that would significantly relax restrictions on the collection of personal information. This bill is currently under deliberation in the House of Representatives.

The bill would allow AI operators and other parties to freely collect information, such as medical history and political beliefs, without the individual’s consent, if such information has been made public on social media or other platforms.

Is there any risk that AI could disseminate information such as one’s medical history? Could individuals be placed on certain lists based on their political beliefs and suffer discrimination as a result? One cannot help but harbor concerns about the government’s stance of prioritizing AI development. This matter should be deliberated carefully in the Diet.

(From The Yomiuri Shimbun, May 10, 2026)