On September 16th, the Humanitarian AI+MERL Learning Group hosted the official launch of “Aid and ID: Making People Matter”. Alongside editor Margie Cheesman, we heard from book contributors Jaki Mebur, Ryan Beech and Bronwen Manby.  Together we discussed how identity systems can be designed for the end user, the possibilities for ID systems to provide greater agency to those receiving aid, and the importance of keeping things analogue in a digital time. 

As technology advances, ID remains essential to humanitarian work

“Throughout the technological developments we learn that even as humanitarian management has become more and more dataified, complex and systemic, the everyday objects —these material things such as ID cards and hard drives— remain essential at the frontlines.”

So much of the humanitarian discussion has, in recent times, been dominated by emerging technologies. Blockchain, drones, and generative AI have swept the sector in successive waves. No stranger to these newer tools, editor Margie Cheesman emphasised the significance of continued attention to foundational tools like identity. She noted how the book takes stock of the past and presents risks and challenges through practitioner experiences, offering insights into opportunities for shaping the future of identification. 

Margie noted three intersecting themes highlighted by contributing practitioners in Aid and ID:

1. Aid recipients often remain enclosed in closed-loop, short-term functional systems. 

Ongoing struggles to build bridges between functional (humanitarian-led) and foundational (national) systems have meant that even as aid recipients might receive identity documents, they often remain excluded from mainstream social and economic life. 

2. Differing mandates of humanitarian organisations result in duplicative ID work.

Heightened efficiency pressures mean that the sector can’t continue in this fragmented manner, but complexities around coordination, control and competition are hindering attempts towards interoperability. 

3. Independent control of humanitarian systems is currently being squeezed. 

As private actors take on even bigger roles in humanitarian work, and more states develop their own identity infrastructures, the status of humanitarian independence is increasingly vulnerable. Alongside these patterns, competing notions of sovereignty for states and the humanitarian sector raise questions about the threshold of independence that the sector might be able to maintain. 

Threaded through these concerns is an open question about how to localise, especially when a lack of funding both fuels the push for centralised, outsourced identity systems while reducing the ability to support participation and inclusion of local organisations and affected communities themselves. What does the future for humanitarian identity hold amidst these intensifying pressures?

Learning from those who do: practitioner insights on identity

Our three book contributors gave lightning talks on their chapters, highlighting key takeaways and offering insights to inform future design of identity systems.

“A human centred approach is my gospel when it comes to digital design.”

Jaki Mebur (Lead, Digital Product Design and Development Portfolio at the International Federation of Red Cross and Red Crescent Societies) shared insights on designing identity programs, drawing on specific learnings from implementing systems in Somaliland. In particular, she noted the need for thoughtful design approaches to ensure inclusive identity practices. Her experiences foreground the importance of human-centred design: specifically, centring the experience of those receiving aid. Institutional tendencies to make design choices related to organisational and even team-level preferences can result in repetitive processes that can cause frustration for aid recipients. Prioritising dignity in digital design requires that institutional silos be broken down to facilitate dignified digital solutions.

“The next big paradigm shift for humanitarian identity management […] is really giving the people themselves that we serve more agency over their information and participation in our programs.”

Ryan Beech (Lead, business operations teams for WFP’s Identity Management and Delivery) offered strategies for giving individuals more direct access and control over their information. Self-registration, for instance, provides one way to reduce the time and costs associated with registration—an approach he argues can also empower individuals to take action faster. Although this approach may not be immediately possible to implement everywhere, Ryan suggested that working towards models of self-ownership should inform how we think about and design registration processes.

Humanitarians are not the only stakeholders that have grappled with identity management. Indeed, as Bronwen Manby (expert on statelessness, nationality and identity documents, affiliated with LSE, School of Advanced Study, University of London and the European University Institute) pointed out, “identification is a necessary tool of any bureaucracy […] some form of list keeping is necessary— the question is how those lists work.”  In reflecting on the book, Bronwen identifies several key tensions among the practitioners that contributed. Varying ideas of efficiency, proportionality and dignity produce different assessments on whether or not biometrics and other technologies generate efficiencies and, if so, at what cost? Not only is the impetus for biometric use debated by different contributors, but also the comparative risks of centralised and decentralised systems and the appropriate approach to data sharing with host states. 

How do we make people matter?

Our speakers reflected on the everyday labor of humanitarian systems, and how practitioners handle and deal with the dozens of small decisions that go into shaping identity systems. Hearing first-hand from those who have made critical design choices, or shaped the decision-making environment of identity, can inform efforts to successfully build policies and shared standards around identity. Margie noted that, in thinking of the humanitarians actively making these decisions, we can understand better how they navigated operational constraints and share important lessons learned. Through this discussion, it is also possible to pinpoint unresolved questions that the sector will need to grapple with in further iterations of identity policy.

Identity decisions and questions of humanitarian sovereignty are intertwined

As a fundamental building block of the humanitarian sector, identity management is reflective of some of the most persistent humanitarian challenges. Decisions about vendors, as well as data sharing and governance, are also decisions that reflect concerns about ownership both among humanitarian organisations, and between the sector itself and private providers. As humanitarians try to design tools that they own, and that correspond to specific humanitarian needs, Brownwen reiterated the need for the sector to come up with not only shared identity standards, but also collective governance standards. 

Rethinking the necessity of registration and the possibilities of alternatives 

Becoming more discerning about what services require registration can also help centre dignity. Jaki probed the conditions under which it may be possible to deliver humanitarian assistance without registration, while Ryan discussed the importance of reducing repeated identification requests. Good enough authentication—more akin to the experiences of signing up to non-humanitarian services demonstrates an existing precedent of proportional identification requirements. Minimising registration aligns with principles of data minimisation, and can re-centre the dignity of recipients by reducing onerous registration requirements. Although the sector already tries to adapt registration practices to different contexts, they must contend with funding expectations and programmatic expectations that may expand the boundaries of data collection. As one speaker noted, this may be possible for short term interventions but is likely to become more complex as programs become lengthier.

Reducing friction in identity systems can transfer responsibility 

Lastly, amidst the discussion of greater agency and the promises of self-registration, it is important to remember that self-registration systems also require work. As Margie noted, self-registration not only transfers the effort of registration to recipients, it may also externalise data-sharing responsibilities to them. One remedy proffered by Jaki is for humanitarian organisations to do the institutional work to make it easier for those receiving assistance and reduce friction for them alongside efforts to provide greater agency. More fundamental agreements about what identity is and how identity systems can be constructed in a human-centred way are an important step towards addressing both humanitarian paternalism and the responsibilisation of recipients. 

Q+A

We were so lucky to enjoy a rich discussion and tackle many difficult questions. Here’s a snippet!

• On first principles…Although there are currently no overarching principles that all humanitarian agencies adhere to, organisations have their own processes that are at times informed by other shared principles—GDPR being the most common. Speakers reiterated that underlying most of this work remains the humanitarian principles, in particular the principle of dignity and the commitment to ‘do no harm’
• On pursuing interoperability while avoiding the introduction of new harms…In a rapidly changing landscape, it can be difficult to establish technical standards that can provide consistency. Centering the right to privacy, however, can help with focused humanitarians in the design of new interoperable tools and standards. Additionally, it may help to consider the analog nature of many of these challenges and look to the ways the humanitarian principles already function as first principles when designing standards and, by extension, designing systems.
• On dealing with the inescapable shadow of AI…Our speakers largely agreed that the need for humans does not go away with the introduction of AI. Especially in the case of identity systems, speakers highlighted the specialist and very human skills needed to help people figure out how to properly navigate system requirements. The complexity of identity service delivery means that it is even more important that those interacting with humanitarian organisations can speak to a person when they need help. 
• On building consent…The IFRC has built a tool that layers consent and data sharing together according to the services being sought by an individual. Additionally, the tool is written in plain language to ensure users can actually understand—and thereby consent—to data sharing. 

The human touch remains essential to making systems work

As Margie reminded us, identity systems are not secondary concerns for aid recipients—in fact, they are often actively pressing upon how they experience the day-to-day of humanitarian work. All three book contributors reinforced the importance of offering face-to-face methods of registration, appeals, and assistance even as digital tools become ever more central to registration. Making room for the human aspect of registration requires funding, however, and all our speakers noted the importance of continuing to financially support efforts to have humanitarians on the ground. The continued relevance of analog remedy and communication seems to be more important now than ever before.

Further resources

Thanks to all those who joined the call. We are especially grateful to the book contributors for their time and insights. If you would like to watch the recording or revisit the notes, you can do so here.

And if you are interested in getting a physical copy of the book, you can do so here. 

If you have ideas on speakers, topics of discussion, or any other ideas, please reach out to our working group lead, Quito Tsui. And if you haven’t already, find out more about joining the NLP Community of Practice here.