FDN - Our Approach
Frontier data technologies and AI offer immense potential for improving the lives of children, but they face significant barriers to adoption. The Frontier Data Network takes a values-driven, ecosystem approach that addresses these interconnected challenges holistically. Our work transforms how data and AI serve children in the most vulnerable situations worldwide. FDN believes in the need for an ecosystem of partnerships that can create opportunities to innovate and ensure impactful solutions globally.
CHALLENGES FOR DATA, TECHNOLOGY, AND CHILDREN
ALGORITHMIC EQUITY, DATA BIAS, AND DIGITAL DIVIDE
Frontier data technologies are becoming essential to decision-making in humanitarian and development contexts. However, accessing new data streams remains a major challenge. Even when access is possible, the digital divide often leads to bias, underrepresentation, and algorithmic unfairness, resulting in decisions that overlook the most vulnerable and deepen existing inequalities.
“It is critical to ensure that new data science echnologies and methods are developed based on the needs of the most vulnerable: with more attention to data bias, algorithmic equity, and the ‘last mile’ of humanitarian and development efforts.”
MANUEL GARCÍA-HERRANZ
Principal Researcher of the Frontier Data Network, UNICEF
TALENT SHORTAGES
A critical shortage of skilled professionals in frontier data limits its impact in humanitarian contexts. With an estimated demand for 3.5 million data scientists in the social sector over the next decade, competition for talent is intense and largely driven by the private sector.
LEGACY TECHNOLOGY INFRASTRUCTURES
Integrating advanced data technologies into existing humanitarian and development programs is challenging due to outdated systems and processes designed for a previous era. These legacy infrastructures are often ill-suited to support modern data solutions. Addressing this requires not only closing skill gaps and resolving ethical issues but also investing in digital transformation and infrastructure upgrades to enable frontier tech projects and talent to thrive.
Our approach to these challenges is built into the FDN’s core strategy: applying frontier data technologies and Artificial Intelligence (AI) in critical use cases, building modern data architectures, and building UNICEF’s capacity for data-driven decision-making. This holistic framework addresses immediate barriers while laying the groundwork to fully harness cutting-edge technologies—ultimately enhancing our ability to respond to crises and improve the lives of children around the world.
AN ECOSYSTEM APPROACH
1) ADVANCE THE USE OF FRONTIER DATA TECHNOLOGIES AND AI IN CRITICAL AREAS OF APPLICATION
By advancing the use of Frontier Data Technologies (FDT) and AI in critical application areas, we directly address urgent humanitarian and development needs through innovative solutions. Investing in these targeted applications ensures that data technologies are deployed where they can have the most immediate and significant impact for children in crisis situations and vulnerable contexts.
IMPACT
Targeted application of advanced data technologies in crisis settings enables faster, more effective humanitarian responses that directly improve outcomes for children. These applications generate use cases that demonstrate the real-world impact of data-driven approaches, creating models that can be replicated and scaled across different contexts. By focusing on critical areas with high urgency, we maximize the potential for transformative impact while building evidence for broader adoption of these technologies.
EXAMPLES
Ukraine and Gaza Geospatial Support
Utilizing high-resolution satellite imagery and geospatial analysis to support research and humanitarian efforts in conflict-affected areas. These interventions leverage advanced AI techniques to rapidly assess damage, identify critical infrastructure needs, and guide humanitarian response activities in real time, directly improving outcomes for affected children.
Migration & Environmental Emergencies
Developing data technology solutions to address the unique challenges faced by children on the move and those affected by environmental disasters. This includes implementing frontier data approaches to better identify, track, and respond to migration patterns and climate-related emergencies, ensuring that vulnerable children receive timely and appropriate support during crises.
2) MODERN DATA ARCHITECTURE
Investments in digital data platforms and systems ensure that UNICEF and its partners can overcome the limitations of legacy technology infrastructure and harness robust, scalable systems to support the best available data science and analytics practices for humanitarian programmes.
IMPACT
With modern data architecture, we can provide real-time insights and data-driven solutions that are crucial for responding swiftly and effectively to crises. Facilitating access to internal and external datasets enhances humanitarian decision-making capabilities, making interventions more targeted and efficient. Establishing a robust data infrastructure allows us to leverage external expertise and resources, further strengthening our ability to harness the power of big data and advanced analytics. Ultimately, these efforts improve the lives of children worldwide by enabling more informed and impactful humanitarian responses.
EXAMPLES
Geospatial Analysis Infrastructure
Developing open-source geospatial platforms that enable UNICEF staff and partners to visualize and analyze location-based data with minimal technical training. These platforms include GeoSight for creating online maps and GeoRepo for storing and sharing standardized administrative boundaries and other geospatial datasets, supporting the work of both UNICEF staff and external collaborators.
Global Immunization Budget Database (G-IBD)
Leveraging AI technology to consolidate and standardize unstructured government budget data across countries. This platform provides an easy-to-use interface to analyze immunization budget data at country, regional, and global levels, helping decision-makers allocate resources more effectively and monitor healthcare spending for children.
Additional Examples
- Indicator Data Warehouse: A comprehensive data repository provides access to various datasets crucial for monitoring progress towards SDGs concerning children
- How Many Tool: An AI-powered interface provides data-driven responses to user questions in plain language, removing technical barriers to accessing UNICEF’s data
3) BUILDING CAPACITY FROM FIELD TO GLOBAL PARTNERS
By fostering a culture of data-driven decision-making and strengthening capacity, we can accelerate the integration of modern data technologies in humanitarian and development contexts. The Frontier Data Network is building an interdisciplinary community to facilitate the sharing of knowledge, resources, and best practices, creating a mutually supportive environment for continuous learning and innovation.
IMPACT
Building capacity from the field to global partners ensures that data technologies are not just implemented but are utilized effectively to drive impactful humanitarian outcomes. This approach strengthens local expertise while connecting it to global networks of knowledge and resources. Promoting knowledge-sharing activities among countries participating in priority use cases creates powerful collaborative learning environments. By investing in partnerships with academic institutions, particularly in the Global South, we build an effective global research framework that uses AI and frontier data to strengthen national data systems. This holistic approach enables more informed decision-making and enhanced programme effectiveness.
EXAMPLES
Development Data Partnership for UNICEF
The FDN leads UNICEF’s collaboration with the Development Data Partnership, which facilitates access to 30+ private sector technology data providers, including Google, JBA, and Ookla. This initiative enables data science collaborations to address critical development and humanitarian sector challenges by unlocking access to novel data sources from private sector companies.
Strategic Collaborations
Establishing partnerships with leading academic institutions (e.g., MIT, Harvard, University of Edinburgh) and technology leaders to address talent gaps in critical areas like data science, engineering, and machine learning. These collaborations ensure progress on science-based projects and prevent wasted resources by leveraging external expertise.
Additional Examples
- Data for Children Collaborative (D4CC): Providing expert data science support from the University of Edinburgh for various projects
- Regional Nodes Development: Building capacity in key regions (Latin America, West and Central Africa, and East Asia and Pacific) to address specific regional needs while maintaining global linkages.