AI use in programs and operations
There are four core functional areas in which AI is already enhancing workflows, supported by real-world examples.
Deep research
AI rapidly scans, analyzes, and synthesizes large volumes of academic papers, reports, and complex documents, extracting key findings, summarizing content, identifying patterns, and structuring insights. Tools like Microsoft Copilot have deep research modes that support this usage. Dignity means respecting every person and keeping their information safe. Nonprofits often collect sensitive data, so any AI system must handle it carefully.
Data aggregation and analysis
AI can analyze large datasets, surface patterns, generate real-time insights, and visualize information through tools like Power BI and Copilot in Excel. These tools reduce the time needed for manual data processing.
Case study
Africa Infodemic Response Alliance case study
Africa Infodemic Response Alliance (hosted by WHO) uses AI to process vast streams of online and community-generated data. This process detects misinformation and public concerns, demonstrating how AI supports large-scale data aggregation and insight generation.
Monitoring, evaluation, accountability, and learning (MEAL) dashboards
AI supports MEAL workflows by generating indicators, summarizing performance data, and automatically updating dashboards. It helps teams identify anomalies and track changes over time. Power BI and Microsoft Foundry are common tools used for MEAL dashboards and text analysis.
Case study
fAIr, Humanitarian OpenStreetMap Team (HOT)
fAIr (HOT) uses AI models, specifically with computer vision techniques, to detect objects in satellite and unmanned aerial vehicle images. This allows for rapid identification and mapping of buildings and roads, strengthening geospatial insights available for MEAL and program planning. Use of AI has doubled the speed of mapping efforts.
Predictive analysis and anticipatory action
AI models combine hazard forecasts (flood predictions) with local data to trigger early program responses, such as anticipatory cash transfers or targeted interventions, before a shock occurs. Microsoft Azure Machine Learning can be leveraged for this.