Microsoft: G7 Hiroshima AI Process (HAIP) Transparency Report
Summary
Microsoft's transparency report regarding its AI risk identification, evaluation, and management practices as part of the G7 Hiroshima AI Process.
Key quotes
We have defined “tracked high-risk capabilities,” which include: Chemical, biological, radiological, and nuclear (CBRN) weapons... Offensive cyberoperations... Advanced autonomy.
This report details Microsoft’s internal governance frameworks, including the Frontier Governance Framework and the Responsible AI Standard. It outlines their approach to red teaming, automated measurement, and compliance with international security standards.