Back to news

Collective crisis intelligence for frontline humanitarian response

Authors: Aleks Berditchevskaia, Kathy Peach, Isabel Stewart

Two linked Nesta reports analyse how collective crisis intelligence (CCI) improves anticipation, management and response in the humanitarian sector, and document the results of a project to design and evaluate new CCI tools.

Two linked Nesta reports analyse how collective crisis intelligence (CCI) improves anticipation, management and response in the humanitarian sector, and document the results of a project to design and evaluate new CCI tools

Collective crisis intelligence for frontline humanitarian response

The first report analyses how an emerging innovation approach, ‘collective crisis intelligence’ (CCI), is being used to improve anticipation, management and response in the humanitarian sector.

CCI combines artificial intelligence (AI) with methods that gather on-the-ground human intelligence from communities affected by crises and frontline responders.

The report is aimed at the wider community of humanitarian innovators and innovation funders, with the hope of spurring research and development (R&D) in this emerging area.

As the first landscape analysis of CCI solutions in the humanitarian sector, the research presents a technical analysis of existing solutions, looking at a range of dimensions that include data sources and technology types.

Key findings

1. CCI is mainly used for early warning of crises and real-time information for effective response

More than two thirds (68%) of CCI solutions are being used to provide early warning of a crisis, or real-time situational information during the preparedness and response phases of crisis management.

Almost half of all cases focus on rapid-onset disasters, such as floods, earthquakes and hurricanes.

2. Collective crisis intelligence is a nascent field

Many of the CCI solutions analysed are at an early stage of development, with a significant proportion in concept/idea or prototype stages.

Their early stage of development means that few have been integrated into humanitarian workflows or systems.

Better integration is needed, which will require work to overcome organisational and technical barriers, such as lack of leadership buy-in and digital skills gaps.

3. CCI could strengthen localisation, anticipatory action and more human-centred AI 

By drawing on novel data sources, including from responders and communities on the frontline, CCI solutions could build a richer local and social understanding of crises.

Combining these solutions with the processing power of AI technologies would give humanitarians access to more timely and contextual data, which could be used for anticipatory action, effective response or sustainable recovery.

4. Emerging applications of CCI include modelling interventions for more effective programme planning

CCI solutions could better support longer-term planning and decision -making for mitigation or recovery efforts through modelling interventions.

This could enable different stakeholder groups to gain a collective understanding of impacts, dependencies, and emergent or unintended effects, enhancing collaboration and building trust.

Ten R&D opportunities for CCI

Expanding CCI solutions to new users

  • Developing CCI solutions that draw on the expertise of frontline workers and enable them to take action.
  • Using CCI methods to deepen community participation through active (rather than passive) contributions, such as crowdsourcing ideas.

Applying CCI solutions to new issues in crisis management

  • Expanding situational awareness of misinformation and disinformation through CCI solutions.
  • Predicting the resources needed for crisis mitigation, response and recovery.
  • Involving communities in real-time monitoring and evaluation of humanitarian response and recovery efforts.
  • Leveraging CCI for distributed intelligent actions that involve traditional and non‑traditional actors.

Leveraging new technologies in CCI solutions

  • Leveraging unsupervised or semi-supervised machine learning techniques for improved situational awareness.
  • Modelling the complexity of crises and the effects of humanitarian challenges and actions.
  • Using participatory modelling for improved multi-stakeholder decision -making.
  • Using CCI to bridge the gap between human reasoning and AI predictions.

The research was part of a larger project on CCI for humanitarian action that Nesta is delivering in partnership with the IFRC Solferino Academy.

Localising AI for crisis response – putting power back in the hands of frontline humanitarians and local communities

The second report documents the results of a year-long project to design and evaluate two new proof-of-concept CCI tools.

The tools combine data from communities affected by crises with the processing power of AI to improve humanitarian action:.

  • NFRI-Predict anticipates which non-food aid households in different regions of Nepal will need after a crisis.
  • Report and Respond is an SMS-based tool that allows Red Cross volunteers in Cameroon to check the accuracy of rumours and misinformation about Covid-19 they hear from the communities they work with, and receive real-time guidance on appropriate responses.

Both were developed using Nesta’s participatory AI methods, which aim to address risks associated with humanitarian AI by involving local communities in the design, development and evaluation of new tools.

The report found that CCI has the potential to make local humanitarian action more timely and appropriate to local needs, and could transform locally generated data to drive new forms of (anticipatory) action.

It also found that participatory AI methods could overcome limitations of AI, as well as helping to improve the performance of models.

Such methods could also expose tensions between the assumptions of and standards set by AI gatekeepers and the pragmatic reality of implementation.

They could also create opportunities for building and sharing new capabilities among frontline staff and data scientists.

The report validated the belief that CCI and participatory AI could increase trust in AI tools, but more research is needed to untangle which factors are responsible.

It also demonstrated that it is possible to build AI that responds to local values and priorities using local infrastructure, data and talent.

However, more investment and experimentation is needed to realise a future where locally developed and owned AI becomes ‘business as usual’.

The report was published alongside two technical reports for the tools used in Nepal and Cameroon, which each include a detailed methodology, and GitHub repositories containing the code, prototypes and technical specifications, with recommendations for future development.

The project was a partnership between Nesta’s Centre for Collective Intelligence Design and Data Analytics Practice, the Nepal Red Cross Society and Cameroon Red Cross Society, the IFRC Solferino Academy and Open Lab Newcastle University.

Both projects were funded by grants from the UK Humanitarian Innovation Hub (UKHIH). 

The UK Humanitarian Innovation Hub is an independent entity hosted within Elrha and fully funded by the UK Foreign, Commonwealth & Development Office.