Building a safer AI future in the open
The Open Secure AI Alliance builds and shares an open defensive stack of AI models, tools, and techniques to strengthen cyber defense and secure software and AI agents—giving defenders everywhere capabilities they can inspect, adapt, trust, and control.
Our work is organized into Working Groups, each focused on a specific domain. Working Groups are proposed by any Alliance member and approved by the Steering Committee.
| Working Group | Repository | Status |
|---|---|---|
| SAFE — Shared AI Findings Exchange | wg-safe | Approved |
To propose a new Working Group, see the governance repository.
| Repository | Purpose |
|---|---|
| governance | Alliance charter, WG formation policy, Steering Committee records |
| project-template | Template for creating new Working Group repositories |
| wg-safe | SAFE Working Group — Shared AI Findings Exchange |
- 💬 Join our Slack — introduce yourself in
#welcome - 📋 Read the governance docs — understand how the alliance works
We encourage contributors to work with their employers to become organizational members of OSAIA. Members may participate in any Working Group and have voting rights in Working Group decisions.
All participants in the Alliance spaces are expected to follow our Code of Conduct (link coming soon). We are committed to a welcoming, respectful, and inclusive community.