Joshua Blumenstock states that a humbler data science could transform international development while also limiting the number of alleged silver bullets that have missed their mark in recent years. Describe the promise, pitfalls and ways forward Blumenstock uses as the foundation for his thesis. Additionally, consider the following statements from three of your classmates regarding this article. (1) “Good intent is not enough in data science when dealing with the problems which determine people’s experiences” Anna Raymond (2) “Transparency is the underlying issue to many of these problems, so an increase in this on both ends (data based issues & human based issues) could lead to better results.” Nira Nair (3) “In lieu of such drastic potential for promoting applications yet demoralizing hindrances, the balancing act can become difficult.” Kayla Seggelke How do you respond to these ideas regarding “good intent”, “transparency” and the difficult “balancing act” when considering the intersection of human development with data science?

The promise is that governments and aid organizations can distribute humanitarian aid much more effectively, in a focused and timely way, than is done now. The pitfalls of this approach are that there will be: (i) unanticipated consequences (the people and organizations using big data will only use it to benefit themselves, not for humanitarian efforts, but rather to make profits); (ii) a lack of validation of the algorithms used in parsing through the data collected (examples include whether what happens in practice matches what the data science algorithms predict and whether these algorithms will remain effective over time); (iii) biased algorithms, e.g. the people who most need humanitarian aid or digital credits don’t even own a cell phone or have access to electricity in order to be included in the data set being used in the algorithm; and (iv) a lack of regulation (i.e. for those persons whose data are collected there are no privacy laws or checks and balances to protect their data from being collected or shared with for-profit private companies).

The ways forward that Blumenstock proposes are to: (i) validate digital data collection and the algorithms using that digital data with traditional data collection methods such as face-to-face surveys and benchmarking digital data with other conventional data collection methods; (ii) customize data collection methods to make them more fair, accountable, and transparent (FAT) (e.g., by making an “impact score” in addition to a “credit score” to enable lenders to avoid making “pay-day” type loans to borrowers who many not need them); and (iii) deepen collaboration with data scientists, development experts, governments, and others from the public sector who can help private companies “do good” in addition to just making money using data collected on citizens in third world countries.

Regarding the responses from my classmates–”good intent,” “transparency,” and “balancing act,”–they are all appropriate criticisms of the intersection between human development and data science. As Blumenstock points out in his analysis, oftentimes for-profit, private companies, the main users of data science when making for example digital credits, lack the “good intentions” that governments, aid organizations, and other public sector entities have when rendering assistance in developing countries. To overcome this pitfall, he proposes to deepen collaboration with these entities which I believe would help greatly to generate goodwill and offer greater insight into the problems individuals face on the ground. Transparency is a similar pitfall Blumenstock points out, where he suggests that regulation and government oversight, which exists in the United States and European Union, should apply to data protection and privacy. In turn, I think this would help make the collection and use of individuals’ data more transparent. Finally, I believe the balancing act between transparency and profit, between “doing good” and making money, between collaboration and acting alone as a private enterprise, is a delicate one. However, if data science is to play a role in human development and help overcome humanitarian crises, in my opinion getting the correct balance is crucial. Private companies need to make money implementing data science algorithms (otherwise they will not do it), but at the same time humanity should not be a testing ground for algorithms and data scientists. Humanity is more important than that.