To encourage developers and service providers of generative artificial intelligence (AI) technologies to adhere to the Guiding Principles,JapanThe government will consider new measures to alleviateGenerative AIThese measures may include establishing a third-party certification system and strengthening supervision of industries and organizations deemed high-risk.
Source Note: The image is generated by AI, and the image is authorized by Midjourney
The move comes after the Group of Seven (G7) advanced economies agreed on an international code of conduct for developers of generative AI. The government on Tuesday presented recommendations on how to reduce AI-based risks at an AI strategy board made up of AI experts and government officials.
To ensure that developers take risk mitigation measures and disclose information to increase transparency, the committee is expected to consider conducting external audits and establishing a third-party certification system. The government also plans to hold regular meetings between government officials and AI developers to exchange opinions.
For generative AI service providers, the source said eight sectors were listed as high-risk examples, including government, finance, energy, logistics, transportation, telecommunications, broadcasting, and medical care. For these sectors, the government is expected to coordinate and strengthen supervision of service providers by formulating additional rules.
For low-risk areas, the government plans to require service providers to disclose information by developing governance policies for operating AI systems.
The government plans to study overseas cases and propose specific measures by the end of this fiscal year to reduce the risks of generative AI.
As this year’s G7 presidency, Japan is leading the Hiroshima AI Process, a framework designed to discuss the challenges of generative AI. At the end of October, the G7 agreed on an international code of conduct that calls for appropriate actions to address AI-based risks, as well as the development and introduction of methods such as digital watermarking technology to help users identify AI-generated content.