Page 117 - AI Standards for Global Impact: From Governance to Action

AI Standards for Global Impact: From Governance to Action

in Trinidad and Tobago. LetzFarm is expanding through global data protocols, regulatory
frameworks, localized training for Small Island Developing States (SIDS), and educational
partnerships with universities for AI-based agricultural training and research.
v. WFP highlighted how AI can reshape food systems by shifting from reactive crisis response
to proactive risk management, enabling rural communities and national planners to Part 2: Thematic AI
anticipate shocks like droughts, floods, and market volatility. WFP partnerships with
Google, Microsoft, IBM, and the Gates Foundation have advanced food insecurity
forecasting, crop yield prediction, micronutrient analysis, and geographic targeting. A
key focus is on strengthening data channels, real-time analytics, and scalable AI models,
with successful pilots in Yemen, Nigeria, Cameroon, and Ethiopia. While AI is powerful,
its impact depends on complementing human expertise and being built on trust,
transparency, and ethics to enable faster, more targeted interventions for food security.
vi. The German Agency for International Cooperation (GIZ) and the Gates Foundation
focused on the challenges faced by smallholder farmers, including limited access to
personalized digital advisory services, and the lack of solutions designed for low-literacy,
low-digital skill groups. They outlined the opportunity to leverage recent advancements in
AI and digital technologies to deliver personalized and dynamic information to farmers. A
use case was presented where a farmer named Rose uses a feature phone and Interactive
Voice Response to receive farming advice in her local dialect. The project timeline includes
key milestones from 2023 to 2025, such as the call for proposals, Minimum Viable Product
development, and end-user research and evaluation. Lessons learned include the need
for multi-disciplinary partnerships and further investments in foundational technologies
to sustain agricultural advisory systems.

17�4 Key outcomes

Key outcomes and findings can be summarized as follows:

1) There is a need for transparent AI standards in digital food, with ITU positioned to
collaborate with FAO and other UN partners on global frameworks for data governance
and AI governance toolkits.
2) Gaps were identified with respect to open data protocols, and the development of scalable
digital advisory platforms, particularly for smallholder farmers with low literacy and limited
digital skills.
3) The Digital Agriculture & AI Innovation Roadmap and NaLamKI project highlight
opportunities for ITU to support the standardization of AI interfaces, data-sharing
protocols, and validation of AI-powered digital food systems.
4) Multi-disciplinary partnerships and inclusive solutions will be key for standards to evolve
alongside emerging technologies for resilient and future-ready food systems.
5) During the AI for Good Global Summit 2025, the Global Initiative on AI for Food Systems
was launched. It is led by ITU, FAO, WFP, and IFAD. It aims to drive shared digital
infrastructure, pilot projects, and standards to empower governments and innovators to
deliver real-world impact for resilient food systems. The outputs of this initiative will feed
into key ITU standardization workstreams to support adoption at scale.

105