Hiroshima AI Process Reporting Framework

The HAIP Reporting Framework provides a common, voluntary structure for organisations to disclose their AI governance and risk management practices. It supports the implementation of the Hiroshima AI Process International Code of Conduct and promotes transparency and good practices across jurisdictions.

Featured event: Advancing AI transparency in practice

An official side event of the India AI Impact Summit 2026

Date and time: 17 February 2026, 11:30 AM to 12:25 PM (IST)

Venue: West Wing Room 4B, Bharat Mandapam Convention Centre, New Delhi, India

Overview

Co-hosted by the OECD, Japan’s Ministry of Internal Affairs and Communications, Infosys, and Microsoft, this discussion will explore how to align national and regional policy frameworks with shared expectations for transparency, safety, and accountability.

Anchored in the Hiroshima AI reporting framework for advanced AI systems, the discussion will explore how common reporting practices, risk disclosures, and transparency mechanisms can support both national policy objectives and cross-border trust.  Participants will discuss how voluntary frameworks, such as international reporting templates, assurance mechanisms, public–private collaboration on testing and evaluation, can complement national policies by enabling faster learning cycles, supporting innovation, and fostering convergence where it matters most for safety and rights.

Interested in attending? Register for the event here.

HAIP report cover

DOWNLOAD THE REPORT

How are AI developers managing risks? Insights from responses to the reporting framework of the Hiroshima AI Process Code of Conduct


Rapid advances in artificial intelligence (AI) are reshaping economies and societies, creating significant opportunities while also raising important considerations around the effective governance and risk management of advanced AI systems. Launched in February 2025, the Hiroshima AI Process Reporting Framework is the first international, voluntary tool to help organisations report on their practices compared to the Hiroshima AI Process International Code of Conduct for Organisations Developing Advanced AI Systems. This report presents preliminary insights from submissions by 20 organisations across diverse sectors and countries, examining their approaches to risk identification and management, transparency, governance, content authentication, AI safety research, and the advancement of global interests.