The Global Initiative on AI for Food Systems, is a collaborative global effort led by the International Telecommunication Union (ITU) together with the Food and Agriculture Organization (FAO), World Food Programme (WFP), International Fund for Agricultural Development (IFAD), and other partners to harness the transformative potential of artificial intelligence (AI) and digital technologies across the entire food value chain. It aims to boost agricultural productivity, strengthen food security, and build resilient, sustainable food systems by embedding AI into areas such as precision agriculture, supply chain optimization, and real-time decision support while addressing climate and resource challenges. A central objective is to establish common frameworks, shared digital infrastructure and standards that ensure interoperability, security, and adaptability of AI applications, enabling governments, innovators and smallholder farmers, who produce a significant portion of global food but often lack connectivity and support—to benefit from AI innovation at scale. The initiative supports knowledge exchange, synergistic partnerships, and proof-of-concept deployments to demonstrate scalable, responsible, and inclusive AI solutions that contribute to global food system resilience.
Activity Type AI Tools/SolutionsTrainings/WorkshopsTechnical AssistanceResearch/Reports/AssessmentsPolicy/Regulatory GuidanceAwareness/AdvocacyNetworks/Mentorship/Exchange
Prisma is WFP’s global supply chain operations control tower, designed to improve planning, coordination, and decision making across humanitarian operations. It consolidates data from multiple systems into a single platform, providing descriptive, predictive, and prescriptive analytics to support upstream and downstream supply planning.
Building on this foundation, Prisma integrates embedded artificial intelligence capabilities that significantly enhance accessibility, automate analysis, and provide forward looking insights. Staff can now interact with Prisma as they would with a colleague—asking questions in natural language and receiving clear, data driven answers in seconds. Instead of relying only on technical focal points to extract data, prepare pivot tables, reconcile reports, or interpret complex dashboards, users can now access operational insights directly and much faster.
AI generated executive summaries translate complex analytics into concise insights for decision makers, while automated alerts and predictive models highlight emerging risks such as supply gaps, delayed deliveries, or stock expiries. Prisma also supports scenario planning, enabling users to explore options and understand the operational impact of different decisions in real time.
The initiative improves operational efficiency, reduces costs and losses, and strengthens cross functional coordination . It also saves significant staff time by reducing manual data extraction, reconciliation, and reporting workloads, allowing teams to focus more on judgement, coordination, and decision-making. ultimately enabling more timely and reliable assistance to vulnerable populations and ensuring that resources are used as effectively as possible.
WFP’s Enterprise Language Model (ELM) is an AI-powered corporate search and knowledge discovery capability designed to help staff quickly access operational information across multiple internal platforms, including the WFP intranet, document repository systems, policies, manuals, and other corporate knowledge sources. By enabling natural language queries and surfacing relevant information in seconds, ELM reduces the time spent searching across fragmented systems, supports faster decision-making, and improves access to institutional knowledge across headquarters, regional offices, and country offices. The platform is designed to assist users in discovering and navigating existing information sources while maintaining the appropriate governance, source attribution, and access controls.
Activity Type AI Tools/SolutionsThe FAIR application was developed to centralize appeal workflows previously managed manually across multiple data collection tools. Its objectives include providing a unified dashboard for appeals, enforcing role-based access, synchronizing appeal data between ActivityInfo and SQL Server, maintaining audit trails, and enabling reporting and data export capabilities. Success criteria emphasize complete appeal lifecycle management within FAIR, role-specific data visibility, and on-demand data export. The project explicitly excludes replacing ActivityInfo as the source system and advanced analytics beyond operational reporting. FAIR aims to improve consistency, efficiency, automation, analytics, and trust in appeal management processes
Activity Type AI Tools/Solutions