Q&A: A criticism of the 'innovation narrative'

NAIROBI — The world is facing an unprecedented number of crises. Politics, natural disasters, and climate change are fueling crises at a faster rate than the humanitarian sector is able to mobilize and respond. Now, funding is spread thinly and humanitarians are expected to do more with less. Because of this, the sector often looks to new data and information communication technology innovations to help humanitarian responses, such as implementing the use of drones, biometrics, and mobile technology. But not everybody agrees with how technology has been integrated into humanitarian approaches. These innovations often raise ethical questions about whether the rights and security of beneficiary populations have been properly accounted for in the rollouts of these new programs. Devex spoke with Nathaniel Raymond, director of the Signal Program on Human Security and Technology at the Harvard Humanitarian Initiative, about some of these concerns. The Signal Program on Human Security and Technology, which was founded in 2012, works to create best practice in the field of humanitarian use of ICT and digital data. Rather than creating more pilot programs, humanitarians should work to create minimum technical standards for technological innovations, Raymond said. The conversation has been edited for clarity and length. You’re not a fan of the “innovation narrative.” Can you explain your criticisms of what you’ve referred to as the “innovation industrial complex”? My concerns about the current focus on humanitarian mitigation are that it has really created and magnified three problems. One is that we began engaging in the experimental use of technologies on highly vulnerable populations without first translating what our duty of care is to protect their rights, security, and dignity. Additionally, we did not begin the current “humanitarian innovation era” thinking about how information and data itself are humanitarian needs for populations — equal to food, water, and shelter. We have been working without a net, in terms of the type of humanitarian doctrine and practice standards, that we have for other fields. The second issue is not having asked the right questions about whether our job is innovation. Whether humanitarians are supposed to be engaged in research and development. I think our job is to absorb technology into our workflows. While that’s related to innovation, it takes different competencies, capacities, and capabilities than the ones that are often currently funded. It takes more management science than it takes technological expertise. What we see from working with U.N. agencies and NGOs in the field is, they need help integrating existing technologies in a responsible and effective way into operations, based on need rather than developing new pilots. The third big issue is really the question of what I call “civilian corporate space,” or “humanitarian corporate space.” How do we maintain our independence, our impartiality and neutrality, when we are dealing with partners. Corporations have a different duty than we do. Their duty is a fiduciary one, often to their shareholders, whereas our duty is to meet humanitarian need and uphold the dignity of populations. Sometimes we can partner without a conflict on those issues, and sometimes we can’t. Humanitarians need clear, bright lines about how we work with corporate actors. I think there hasn’t been that guidance. Do you have any examples of instances where a new innovation was implemented without taking into account the local context? A really good example is Sean Martin McDonald’s paper: “Ebola: A Big Data Disaster” about what happened with the use of call detail records — basically phone records from telecommunications companies — during the Ebola outbreak. The experimental use of technology for contact tracing of the spread of Ebola, without a review of privacy and human subjects research laws internationally or domestically in West Africa. Technically, that operation may have violated, according to McDonald, local and international laws, as it relates to data privacy and human subjects protections. There was an issue where the cell phone towers used in developing the model in Korea were significantly closer together and higher in number than in West Africa, where the towers were further apart and fewer. Basically, the desire to do good with data came into play without a review of the infrastructure, the laws and the ethics, and the actual viability of that data model. There’s also data deluge, where organizations are being operationally paralyzed by overwhelming large amounts of digital data flows. The second is data disparity, where certain populations have more data or more connectivity than other populations. That means that we are seeing one population more clearly than another. Data distortion is a third type of data disaster, where we can get an operational picture that is not accurate because we are using the data wrong, or the data is not appropriate for the purpose. The fourth, and most important, is data damage. This can occur even if all of the technical aspects are right, where we use data in a way that causes harm to a population. It increases their vulnerability and susceptibility to targeting by bad actors, by human rights abuse perpetrators. Or it erodes trust with the population because they don’t think we can safely handle their data, or they don’t understand why we are collecting it. We are at a moment where we are focused a lot on the possibility and the promise of technology, but we have many different ways in which we can undermine what we are trying to do for affected populations. This can sometimes cause secondary data-related emergencies, as we are responding. How does decreased funding available to the humanitarian sector play into the tendency for the sector to want to present new innovations? Preceding the World Humanitarian Summit in 2016, and around the Sustainable Development Goals, there has been a lot of money available from donors for humanitarians to try new tech and to engage in prototypes and pilots. This, in many ways, has created a sort of political economy that incentivizes pilot and prototype approaches. There has been more money available for one-off tests than there has been for building the capacity needed to responsibly and effectively manage new workflows that are changing because of tech. “Humanitarians need clear, bright lines about how we work with corporate actors.” --— Nathaniel Raymond, director of the Signal Program on Human Security and Technology at the Harvard Humanitarian Initiative There are some good news examples of organizations such as the vulnerability analysis and the mapping unit of the World Food Programme. They have thought very hard at how they take these technologies and innovate them into an agency’s workflow by developing new professional capacities, both technically and ethically, to govern their use. That’s really important for organizations increasingly using biometrics and digital IDs that can create new pathways for harm and exploitation for populations. Mercy Corps is another good example of an organization beginning to think beyond the tech itself and how they responsibly absorb the tech into their workflows. But the amount of money available for those types of activities, versus the amount of money available for prototypes, has been disproportionately focused on the one-off, often corporate-NGO partnership. What steps should organizations take to professionalize innovations in a way consistent with humanitarian principles? To treat information communication technologies correctly, they must not be about innovation; they must first be about protection. We have some resources; from the Signal program, the Signal Code, which includes our rights-based approach to humanitarian information use in crisis contexts. I think the critical thing that organizations need to do now is to create a space for convening, to work together towards minimum technical standards.