The International Rescue Committee is deploying artificial intelligence tools, including an AI mentor and satellite imagery pilots, to improve the efficiency and reach of humanitarian aid. Chief Research and Innovation Officer Jeannie Annan stated that humanitarian organizations must be strong advocates and frontier users of this technology while leading on risk mitigation.

The organization uses an AI mentor called aprendIA to reach teachers via WhatsApp. The estimated cost of the aprendIA program is $0.20 per child. In Nigeria, the program is projected to reach nearly one million children by the end of 2026. By 2028, the initiative is projected to reach 6.9 million children.

The International Rescue Committee is piloting the use of AI and satellite imagery to locate children not recorded in census data. This tool is projected to identify 10% to 20% more children than current methods allow. It is also projected to reduce the cost per child by 25%. Pilots for this tool will launch in Somalia in the next quarter.

To further streamline operations, the International Rescue Committee is testing photo-to-digital capture technology to replace manual data transcription. These technological adaptations occur as humanitarian aid funding has decreased by 25% since 2024.

Approximately 260 million people require humanitarian assistance, and over 118 million people are forcibly displaced. Globally, 43 million children under five suffer from acute malnutrition. The International Rescue Committee's simplified treatment programs for malnutrition are 20% cheaper than standard treatment mechanisms.

The organization operates a digital information service called Signpost, which has reached 20 million people. Nearly 591,000 people received one-to-one guidance from moderators through the service. The International Rescue Committee tested its Signpost AI assistant for six months in Greece, Italy, El Salvador, and Kenya.

During testing, the pass rate of the Signpost AI assistant increased from 52% to 77%. Staff reported working approximately 70% faster when using the assistant during this period. However, the Signpost AI assistant failed to provide acceptable answers for approximately 25% of queries during testing. Every response from the Signpost AI assistant is reviewed by a human before being sent to a client.

"We need to be all in on AI. And we need to lead, not follow," Annan said.

Timeline

On October 2, 2026, Jeannie Annan stated that humanitarian organizations shouldn’t only be among AI’s biggest skeptics. We should also be among its strongest advocates; its frontier users; and its most active learners. On the same date, she said we need to be all in on AI. And we need to lead, not follow, on mitigating AI’s risks.

Why It Matters

The deployment of these AI tools addresses a critical gap in resources as humanitarian aid funding has decreased by 25% since 2024. With approximately 260 million people requiring assistance and over 118 million forcibly displaced, the scale of need outpaces traditional funding models. The International Rescue Committee aims to use technology to maintain service levels despite these financial constraints.

The global context involves a number of conflicts that is higher now than at any time since the Second World War. This instability contributes to the 43 million children under five suffering from acute malnutrition. By reducing costs through tools like satellite imagery and simplified treatment programs, the organization seeks to extend its reach to those not recorded in standard census data or served by existing mechanisms.