Risk and mitigation strategies for using AI in humanitarian response (beta)

Papua New GuineaPacificGlobalKenya
Written By Tools & guidance

This is a beta version of this how-to note. Given the emerging and innovative nature of AI in humanitarian aid, we are asking people to test the SAFE AI products and give us feedback ahead of the final development of the SAFE AI framework. Please provide feedback via this form, or email info@cdacnetwork.org before 31 October 2025.

How can humanitarian organisations use AI safely and effectively in crisis response?

This factsheet from the SAFE AI Project identifies the risks you need to be aware of when deploying AI in crisis response, such as eroded trust or misdirected aid, and provides practical mitigation strategies for protecting communities from adverse harms.

This provides considerations for humanitarian actors seeking to develop ethical AI solutions that are ethical and safe for crisis-affected communities.

This material has been funded by UK International Development from the UK government.

Tools & guidance
Previous

SAFE AI Checklist: AI onboarding for humanitarian actors (beta version)
Next

Sudan’s information war: How weaponised online narratives shape the humanitarian crisis and response