AI could transform disaster response. How do we ensure it reaches those who need it most?
How do we ensure AI serves the needs of the world’s poorest and most vulnerable? As leading economists consistently note, such an outcome is anything but guaranteed. The phenomenon of market mechanisms alone failing to broadly distribute the fruits of technological advancement is not new or unique to AI. Despite the rise in living standards since the Industrial Revolution, the broad distribution of the benefits depended on regulators, the labour movement, and public investments in social safety nets.
Philanthropic coalitions have also played a critical role in expanding the fruits of scientific progress to communities in need. For example, GAVI, the vaccine alliance, forged a partnership between philanthropy, international NGOs (INGOs), governments, and the private sector to vaccinate more than half of the world’s children who would otherwise lack access to vaccines. Similarly, a needs-based systems approach fueled by cross-sector partnerships can help match advancing AI capacities to the most significant humanitarian demand areas.
At present, most AI investments cater to the interests of higher-income countries. However, a different world is possible, where AI tools are also built and shared according to the unmet needs and aspirations of the Global Majority. In such a world, we’d have large-scale, widely available AI systems to help boost disaster preparedness and response for millions of people, precluding the loss of countless lives and livelihoods every year. The technology that such systems depend on already exists. Humanitarian organisations just need reliable pathways to access it.
A coalition for putting AI in service of humanitarian needs
Inspired by partnerships that successfully channelled the fruits of scientific progress to sectors and communities underserved by the market mechanisms, we’ve taken a coalition-based approach to overcoming the barriers that humanitarian organisations face in the AI ecosystem. Together with our partners, we built DISHA (Data Insights for Social and Humanitarian Action), a platform to leverage AI capabilities for large-scale humanitarian and development needs. Led by UN Global Pulse, the Secretary-General’s Innovation Lab, DISHA brings together humanitarian and development communities, private sector technologists, ethicists, philanthropists, and data providers to co-create AI applications for improving the efficiency and speed of crisis response efforts.
DISHA presents a new vision for the future of AI: one where impactful, community-driven solutions are built and deployed with a bias towards sustainability and scale. Our vision depends on a few critical enablers. First, humanitarian responders need ongoing access to real-time, diverse, community-centred data sources on evolving crises. DISHA draws on insights from the large swaths of data generated by our phones, digital transactions, and satellites. The team then identifies analytical needs that can be met with AI and are commonly shared by many humanitarian organisations – like dynamic mapping of displacement patterns and infrastructural damage – to pool demand and secure access to near real-time data sources on behalf of multiple humanitarian organisations at once, saving time, effort, and resources. To preserve privacy, DISHA analyses the data without moving it outside the provider’s environment, anonymising, aggregating, and sharing it with authorised user organisations to complement on-the-ground needs assessments. While DISHA is currently focused on humanitarian preparedness and response, its model is replicable for other humanitarian and development domains.
Public infrastructure to enable the success and scale of AI-for-good initiatives
Next, robust public infrastructure will play a critical role in supporting ongoing access to data and facilitating the transition from pilots and prototypes – where too many AI-for-good solutions often stop – to products and platforms ready for use by stakeholders in the field. Today, most resource-strapped organisations rely on multi-sectoral partnerships and/or philanthropic funding that can be hard to secure. For more initiatives like DISHA to succeed and scale to other domains, we need sustained public investment from local, national, and regional governments to complement the support of private sector and philanthropic donors.
On a global scale, public infrastructure should also include developing internationally agreed upon, interoperable, and actionable standards for risk assessment and mitigation and safe and secure data-sharing practices across borders. The recently updated OECD AI Principles inform present global standards for privacy, information integrity, safety, and intellectual property rights while grounding AI development in the principles of trustworthiness, human rights, democracy, and the rule of law. On the implementation side, the UNESCO Readiness Assessment Methodology, a tool derived from the UNESCO Recommendation on the Ethics of Artificial Intelligence, supports countries engaging with AI by helping them assess their readiness and guiding local capacity-building efforts to design and use AI safely and ethically. Both public sector stakeholders and INGOs can benefit from these tools as they find ways to build and deploy AI solutions in the service of communities.
Finally, successfully meeting communities’ needs depends on the level of engagement by frontline experts and communities in designing AI products and platforms. DISHA follows a co-design process, placing humanitarian and development organisations at the helm of product decisions. This model amplifies the perspectives of those closest to the impacted communities in risk assessment and risk-proofing and integrates community values into the platform’s design, testing, and validation. Critically, DISHA is also committed to ongoing iteration based on expert feedback and demonstrated impact in the field.
With global crises increasing due to worsening climate change and conflict, we need fast-acting, reliable solutions focusing on the communities in greatest need. Advancing AI capabilities is only one piece of the puzzle. Data access, robust public infrastructure, and community-led design ensure that AI progress meets humanitarian and development needs at scale. DISHA is committed to expanding our coalition of partner organisations to acquire new insights, accelerate impact, and replicate our model for other domains. Through shared innovation, ambitious action, and partnership, we have a chance for a more prosperous and peaceful future for everyone.
