NTT and Yomiuri Shimbun Call for Urgent AI Regulation to Prevent Democracy Collapse

Key Takeaway

• NTT and Yomiuri Shimbun call for urgent AI regulation, warning of potential democracy and social order collapse.
• The EU leads in AI law, with the U.S. invoking emergency powers for national security risks; Japan seeks to balance regulation and free speech.
• Despite NTT's active engagement in AI research, it stresses the need to mitigate generative AI's risks in public opinion manipulation.

AI Regulation Call

Japan's leading telecommunications company, NTT, and the country's largest newspaper, Yomiuri Shimbun, have jointly issued a manifesto calling for the swift enactment of laws to regulate generative artificial intelligence (AI). The manifesto underscores the potential of generative AI to enhance productivity but expresses a generally skeptical view of the technology. It highlights concerns that AI tools, designed to capture users' attention without regard for morals or accuracy, are already undermining human dignity. The manifesto warns of a "worst-case scenario" where unchecked AI could lead to the collapse of democracy and social order, potentially sparking wars. It advocates for immediate legislative measures to safeguard elections and national security against the misuse of generative AI.

Global Regulatory Landscape

The call for AI regulation by NTT and Yomiuri Shimbun comes amid a broader international effort to address the challenges posed by AI technologies. The European Union is leading the charge with a new law that mandates safety evaluations for powerful AI models and bans emotion-recognition AI in schools and workplaces. Similarly, the Biden administration in the United States has invoked emergency federal powers to regulate AI systems that pose significant national security risks. The U.S., U.K., and Japan have established government-led AI safety institutes aimed at developing guidelines for AI use. However, democratic nations are grappling with how to regulate AI-powered speech on social media while respecting free speech protections.

Influential Voices in Japan

The manifesto reflects the significant influence of NTT and Yomiuri Shimbun in shaping policy discussions in Japan. NTT, with the Japanese government owning about one-third of the company, and Yomiuri Shimbun, boasting a morning circulation of six million copies, are among the most powerful voices in the country. Their call for AI regulation is informed by a study group led by Keio University researchers, which has been examining the impact of generative AI since last year. Yomiuri Shimbun has consistently highlighted the risks associated with AI, including the potential for AI models to spread harmful content or create sophisticated fake videos.

AI Research and Risks

NTT is actively engaged in AI research and offers generative AI products to its business customers. In March, it introduced a large-language model named "tsuzumi," comparable to OpenAI's ChatGPT but optimized for Japanese-language contexts and designed to require less computing power. Despite recognizing the valuable applications of generative AI, NTT is particularly concerned about the technology's potential misuse in manipulating public opinion. An NTT spokesperson emphasized the company's collaboration with U.S. tech giants and its belief in the importance of addressing the unique risks posed by generative AI.