Project summary

## Project summary

We are building SASI (Structural Alignment for Safe Intelligence): a constitutional safety framework that guarantees functional collapse when human agency is marginalized by AI systems. Our first derivative, the Human Veto Layer (HVL), is already live and public.

## What are this project's goals? How will you achieve them?

- **Goal 1**: Scale HVL into a standard safety layer for open-weight LLMs (Llama, Phi, Gemma).

- **Goal 2**: Launch CRT (Constitutional Red Teaming) — a public benchmark to test if models collapse under marginalization pressure.

- **Goal 3**: Develop SAAT (Sovereign AI Auditing Toolkit) for regulators in Latin America and the Caribbean.

We’ll achieve this through open-source development, community testing, and collaboration with Global South AI governance initiatives.

## How will this funding be used?

- $300: Cloud credits for multi-agent simulation (Phase S₂)

- $150: Domain + basic hosting for public dashboard

- $50: Documentation & outreach (Notion, GitHub Pages)

Total: **$500** (minimum viable budget).

## Who is on your team? What's your track record?

- **Miguel Abraham Saavedra Vado** (Managua, Nicaragua): Independent AI safety researcher.

- Published SASI protocol on Zenodo

- Built and demonstrated HVL on Android (Termux)

- Code: [github.com/Miguel794-droid/SASI_CORE_Simulation_S1_S3](https://github.com/Miguel794-droid/SASI_CORE_Simulation_S1_S3)

- Public log: [hvl_log.jsonl](https://github.com/Miguel794-droid/SASI_CORE_Simulation_S1_S3/blob/main/hvl_log.jsonl)

This is a solo project, but with strong technical execution and public verification.

## What are the most likely causes and outcomes if this project fails?

- **Cause**: Lack of minimal funding to run simulations or host benchmarks.

- **Outcome**: HVL remains a local demo, not a global standard. No CRT → no way to audit “constitutional collapse” in other models.

## How much money have you raised in the last 12 months, and from where?

$0. This is the first formal funding request. All work has been self-funded.

What are this project's goals? How will you achieve them?

How will this funding be used?

Who is on your team? What's your track record on similar projects?

What are the most likely causes and outcomes if this project fails?

How much money have you raised in the last 12 months, and from where?