TOKYO, JAPAN / MENA Newswire / – Japan has released a draft revision of its Artificial Intelligence Basic Plan that calls for wider global cooperation on AI risks. The proposal focuses on cyberattacks, model safety, misinformation and governance of high-performance AI. The Cabinet Office opened public comments on the draft from June 19 through June 23. The review follows Japan’s first AI basic plan, approved by the government in December 2025.
The draft says agentic AI has raised the complexity of risks facing governments, companies and the public. It cites technical failures, unclear responsibility, labor impacts and national security threats. The document gives special attention to autonomous AI-enabled cyberattacks. It says such attacks can find unknown software vulnerabilities and make stronger cyber defense an essential part of trustworthy AI governance.
Japan’s plan calls for closer cooperation with foreign government agencies, AI developers and other relevant parties. The measures cover early alerts to software vendors, faster discovery of vulnerabilities and stronger response systems. The draft also calls for more evaluation of the cybersecurity performance of AI models. These steps form part of Japan’s wider effort to align AI innovation with safety controls.
Cybersecurity risks draw sharper focus
The plan places the AI Safety Institute at the center of technical evaluation work. It calls for model testing, traceability, guardrails and crisis information-sharing systems. The draft also says Japan should collect and share international data on AI vulnerabilities and incidents. It links those tasks to the development of guidelines, technical standards and rules for safe AI deployment.
The draft includes measures against misinformation and disinformation created through misuse of generative AI. It supports technologies that can identify AI-generated content and help verify authenticity. It also proposes public consultation services for people who suffer harm linked to AI. Businesses would receive support to create complaint-handling systems that address AI-related problems in their own services.
Governance plan expands global role
Japan’s draft also sets out international cooperation as a core part of AI governance. It names cooperation with like-minded countries and Global South countries in areas such as human resources, capacity building, governance and infrastructure. The plan also refers to international standard-setting work for AI. It says Japan should help build coordination models for trustworthy AI across countries and regions.
The proposal comes as governments face growing pressure to manage AI risks while supporting research, development and public use. Japan’s draft keeps both goals in the same policy framework. It calls for risk reviews, stronger cyber response capabilities and technical evaluation of high-performance AI. The public comment process gives citizens and organizations a formal channel to respond before the plan moves ahead.