**SAGE — the Sovereign AI Governance Engine** is an open-source, agent-driven policy platform for governments. It is led by **Emad Mostaque** and sits inside his broader Intelligent Internet project. The pitch: use AI agents, policy simulators, sovereign compute, and multilingual copilots to help states design, test, and adapt governance frameworks — to create policy *"at the speed of AI."*
## The lineage
The clean throughline: **Emad Mostaque → Stability AI / Stable Diffusion → Intelligent Internet → SAGE, the Sovereign AI Governance Engine.** The official one-pager frames SAGE as a move away from *"document-driven, static, and reactive tools"* toward **governing with AI**.
## Key modules
| Part | What it does | | --- | --- | | **AI-powered policy simulation** | Forecasts the impact of policy decisions before they are implemented. | | **Sovereign compute** | Lets nations use large-scale compute while keeping data sovereignty. | | **Multilingual copilots** | Gives government users recommendations grounded in local law, culture, and administrative context. | | **Open governance app store** | Allows modular, sector-specific governance applications. |
## Inside the Intelligent Internet
Intelligent Internet is Emad's post-Stability AI project for open, auditable, sovereign AI infrastructure. Its whitepaper talks about **"governance failures"** in AI, **sovereign AI agents**, **proof-of-benefit**, **Common Ground** coordination, and **auditable datasets and provenance**. SAGE is the governance-facing application of that stack.
## Governing with AI, not governing people
SAGE is less *"a model to govern people"* and more a proposed **AI-assisted governance stack**: agents and simulations for policy design, regulatory analysis, and public-sector decision-making. The human policymaker stays in the loop; the engine accelerates the design, testing, and adaptation of frameworks around them.
## Sources
