Pilot for the Hiroshima AI Reporting Framework 2.0

Organisations developing advanced AI systems can participate by submitting a report. By sharing information, they will facilitate transparency and comparability of risk mitigation measures.

OverviewAboutHAIP BrandFAQs

The Hiroshima AI Process Reporting Framework, launched in February 2025, provides a standardised, voluntary, and adaptable structure for organisations to report on their alignment with the Hiroshima AI Process International Code of Conduct for Organizations Developing Advanced AI Systems. The first round resulted in 25 reports, analysed in the OECD 2025 report "How are AI developers managing risks?" Based on stakeholder feedback during the first year of implementation, a streamlined version was developed, which this pilot aims to test and refine.

Organisations developing or deploying AI systems are invited to participate by submitting a report. The pilot runs from 27 March to 22 April 2026. Pilot submissions will not be made public and will be used solely to refine the framework . Responses can then be reused for the next reporting round. Organisations may therefore wish to retain a copy of their answers. Reports should not include confidential or proprietary information.

Participate in the pilot

Share your organization's practices through our user-friendly interface. You can save your work in progress and return to complete it later.

Browse 2025 reports

Browse submitted reports to learn about organizations' approaches and practices for safe and trustworthy AI development.

Have questions?

Find out more about the project and how to submit your own report

The Reporting Framework, launched on February 7 2025, is a direct outcome of the G7 Hiroshima AI Process, initiated under the Japanese G7 Presidency in 2023 and further advanced under the Italian G7 Presidency in 2024. It builds on the Hiroshima AI Process International Code of Conduct for Organizations Developing Advanced AI Systems, a landmark initiative to foster transparency and accountability in developing advanced AI systems. At the G7's request and in line with the Trento Declaration, the OECD was tasked with identifying mechanisms to monitor the voluntary adoption of the Code.