SAFE AI – Standards and Assurance for Ethical AI
Making AI use visible, governable and independently verifiable.
SAFE AI turns global responsible AI principles into practical controls, evidence and accountability across the AI lifecycle. It starts with humanitarian action, some of the environments where the consequences of AI getting it wrong are highest, and the people affected may have the least power to see, question or contest its use.
The SAFE AI Framework is available now and provides the governance foundation for the next step of independent assurance.
START HERE
Choose your path
-
New to AI governance?
Start with the glossary →
-
Ready to implement?
Get the SAFE AI Framework and Tools →
-
Want the evidence?
Read the briefing paper on the governance gap →
-
Represent a funder or institution?
Find out more about us →
WHY NOW
From principles to practice
AI is already being used across humanitarian and development settings. The challenge is not simply to agree what responsible AI should look like, but to make it work in practice. SAFE AI turns principles into practical control points across the AI lifecycle: helping organisations decide when AI should be used, understand and manage risk, involve the people affected, document decisions and build evidence about whether safeguards are working. No organisation should have to solve this alone. SAFE AI is designed as shared infrastructure – creating a common approach that organisations can use now, while building the foundations for independent assurance as AI adoption grows.
THE FRAMEWORK
How SAFE AI works
SAFE AI takes organisations through the four stages of implementing AI in a humanitarian context, each tested against a Decision Gate:
Problem definition and concept
Design
Development
Deployment and monitoring.
-
The SAFE AI Transparency Card
The central governance record against which the system is held to account. It accumulates across the lifecycle of an AI deployment, holding the outputs of every stage, every assessment, and every decision in a single auditable document.
-
Community in the Loop
A governance requirement embedded at every stage of the SAFE AI journey. SAFE AI treats participation of affected communities as a source of governance evidence, with documented influence over the system's design.
-
Grounded in humanitarian principles
Humanity, impartiality, neutrality, and independence shape every decision the Framework asks organisations to make. They are tested at each Decision Gate.
Proportionate to risk
SAFE AI applies three risk tiers. The depth of governance scales to the stakes of the use case, regardless of the size of the organisation deploying it.
Tier 1: Baseline. Internal, advisory AI use. Light-touch governance.
Tier 2: Enhanced. AI shapes operational decisions, with human oversight. Full assessment and assurance required.
Tier 3: High risk. AI directly affects people's access to assistance, protection, or information. Community co-design and ongoing monitoring are mandatory.
RESOURCES
SAFE AI in practice
The SAFE AI Framework
The full governance framework and tools for ethical AI in humanitarian action. Free to use and always will be. Scales to fit any organisation. Available now.
-
New to humanitarian AI? Start here
GLOSSARY
Plain-language definitions to help organisations build better, safer and more effective AI partnerships. No technical background needed.
-
Step by step: how SAFE AI contributes to trustworthy AI
BRIEFING NOTE
How each step of the SAFE AI Framework supports the National Institute of Standards and Technology (NIST)’s seven characteristics of trustworthy AI.
-
Risks in humanitarian AI
BRIEFING NOTE
A quick briefing on the risks associated with humanitarian AI, and how the SAFE AI Framework and tools respond to each.
-
The governance gap in humanitarian AI: addressing the structural gap between global frameworks and operational reality
BRIEFING PAPER
The analytical foundation for SAFE AI – establishing the nature and scale of the structural governance gap, and why individual agencies cannot fill it alone.
-
From experimentation to engagement
ACADEMIC PAPER
Evidence on the paradox of participatory AI and power in contexts of forced displacement and humanitarian crises.
-
Co-designing AI solutions with crisis-affected communities
HOW-TO NOTE
Practical guidance for meaningful community co-design of AI in humanitarian contexts. -
Co-design vs. user-centred design for AI solutions
FACTSHEET
A clear breakdown of the difference – and why it matters for responsible humanitarian AI.
-
Addressing power dynamics in participatory AI for crisis-affected communities
POLICY BRIEF
Research on the structural challenges of participation and power when AI meets forced displacement.
GO DEEPER
About the Framework
Version: 1.2, August 2026. Supersedes Version 1.1.
Citations:
Framework: McElhinney, H., Mazumder, A., Tjalve, M., Madigan, S. and Spencer, S. (2026). Standards and Assurance Framework for Ethical AI in Humanitarian Action (SAFE AI): A Governance Framework for Humanitarians using AI (Version 1.2). CDAC Network. https://doi.org/10.5281/zenodo.22080937.
Tools & Guidance: McElhinney, H., Mazumder, A., Tjalve, M., Madigan, S. and Spencer, S. (2026). SAFE AI Tools and Guidance (Version 1.2). CDAC Network. https://doi.org/10.5281/zenodo.22079930.
Licence: CC BY 4.0.
Zenodo records: zenodo.org/communities/safe-ai.
Built by: CDAC Network, The Alan Turing Institute and Humanitarian AI Advisory. Stewarded by CDAC Network. Founding architects: Helen McElhinney, Anjali Mazumder, Michael Tjalve and Sarah Spencer.
Beyond the Framework
-
Community of Practice
We also convene an active Community of Practice on AI Governance to advance consensus-building, thought leadership and collective learning.
Join our Community of Practice →
-
Contributors
SAFE AI was developed by founding architects Helen McElhinney, Anjali Mazumder, Michael Tjalve and Sarah Spencer, with CDAC expert consultants.
Read full contributor bios →
-
Our journey to SAFE AI
CDAC’s commitment to community voice in humanitarian response predates AI: it is the foundation of everything we do. These videos trace our journey to SAFE AI.
Watch videos →
-
What we advocate for
Our three key asks of donors, governments and the sector to make humanitarian AI safe, accountable and ethical.
Read our advocacy messages →
This project has been funded by UK International Development from the UK government.
