Becky Faith
Digital systems are critical to development and humanitarian activities, but they can expose some of the world’s most vulnerable communities to unforeseen risks. Recent examples from Afghanistan highlight these dangers.
Also Read
In September, it was reported that the UK Ministry of Defence had been involved in two separate data breaches, potentially compromising the safety of participants in the Afghan Relocations and Assistance Policy programme, which is intended to protect people who had worked for British forces. These breaches followed revelations that the Taliban now control systems for digital identification that were built using international aid. As a former official with the United Nations High Commissioner for Refugees (UNHCR) put it, “The Taliban have been given the keys to the server room.” Moreover, information in the Tazkira, the digital identity cards used in Afghanistan, reportedly can be used to target ethnic groups.
Organisations and governments tasked with providing cash assistance during humanitarian crises recognise the need for legislation to strengthen data protection, but finding staff with both data protection expertise and knowledge of social protection systems is challenging and expensive. As data-driven systems are introduced into aid work, there is the potential for a significant power imbalance between technology providers, governments, and vulnerable populations. And the commercial incentives that drive the involvement of private-sector companies in this domain often are at odds with humanitarian principles – including that the collection and sharing of personal data should not put people at risk.
But aid groups also are guilty of exposing sensitive data. Four years ago, the UNHCR collected personal information – including biometric data – from ethnic Rohingya refugees in Bangladesh and, in contravention of its own policies, shared it with the government of Myanmar to verify people for possible repatriation.
Similarly, a recently released internal audit of the World Food Programme’s (WFP) provision of digital assistance services in Iraq, Namibia, and Nigeria found that none of the projects had processes in place to ensure data protection and privacy. And research from the non-profit Somali Public Agenda revealed that while huge amounts of data are being collected in Somalia by aid groups and the government, the country has no policies in place to ensure that this information is properly handled.
Proponents of digitalisation argue that it will reduce costs. But that may not be true. Unwanted Witness showed that the procurement process for the digital ID scheme in Uganda was conducted with limited public access to tender or contract documents through a “classified procurement” programme, contrary to Uganda’s Public Procurement and Disposal of Public Assets Act, 2003.
Likewise, an internal audit of the WFP’s SCOPE programme, released in May, found that a system had only recently been put in place that could “rigorously account for the costs and benefits accrued from IT projects and investments.” The WFP has been using SCOPE, one of the largest systems in the world for beneficiary and transfer management, since the mid-2010s.
Moreover, this was not an isolated case. In 2019, a group of civil-society organisations signed an open letter in response to the controversial agreement between the data analytics company Palantir and the WFP warning of a lack of transparency in procurement processes between humanitarian organisations and technology companies. The letter suggested that no third-party audits are possible under the current arrangements, making it impossible to do a future cost analysis of this kind of collaboration.
As private-public partnerships continue to grow in the development and aid sector, more research must be done to map the activities of corporate actors to ensure that these relationships are open to scrutiny. Increased transparency will enable donors and policymakers to verify that collaborations are delivering the financial benefits that drive enthusiasm for digitalisation.
The pressure the humanitarian aid sector faces from powerful donors and others to innovate and adopt digital technologies reflects the urgent need to make such assistance more effective. Digital systems can reduce transaction costs and improve the efficiency of aid, for example by providing funds to communities in remote areas via remote electronic cash transfers. But we must guarantee that the world’s most vulnerable communities are not forced to make themselves visible to governments that may not have their best interests at heart.
The writer, a research fellow and Leader of the Digital and Technology Cluster at the Institute of Development Studies, works on the Better Assistance in Crises (BASIC) Research programme at IDS. ©project syndicate
