Work towards responsible information sharing and incident reporting among organisations developing advanced AI systems, including with industry, governments, civil society and academia. Code of Conduct detail: share evaluation reports, security and safety risk information, dangerous intended or unintended capabilities and attempts to circumvent safeguards; establish or join mechanisms to develop and adopt shared standards, tools and best practice; ensure documentation and transparency across the lifecycle, especially for systems posing significant risks; collaborate across the lifecycle to share and report information to the public and relevant authorities; and protect intellectual property in such reporting.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.