# Shared AI Findings Exchange **Aliases:** SAFE **Entity class:** Open AI-security findings-exchange project **As of:** 2026-09-28 ## Definition The **Shared AI Findings Exchange**, abbreviated **SAFE**, is an open project associated with the [[wiki/Open Secure AI Alliance|Open Secure AI Alliance]] for sharing AI-safety and security findings. NVIDIA names SAFE as an example of the alliance's broader work alongside the ecosystem contributions made through the [[wiki/NVIDIA Open Agent Safety Platform|NVIDIA Open Agent Safety Platform]]. The two objects should not be collapsed: the platform is a runtime and reference-system design, while SAFE concerns exchange of security findings. ## Relationships - **alliance:** [[wiki/Open Secure AI Alliance|Open Secure AI Alliance]]. - **adjacent platform:** [[wiki/NVIDIA Open Agent Safety Platform|NVIDIA Open Agent Safety Platform]]. - **domains:** [[wiki/AI Safety|AI Safety]] and [[wiki/Cybersecurity|Cybersecurity]]. ## Sources / Provenance - [NVIDIA Newsroom — Open Agent Safety Platform launch](https://nvidianews.nvidia.com/news/open-agent-safety-platform). - [Open Secure AI Alliance](https://secureaialliance.org/).