Thoughts on a growing debate
Shannon Valor is the Baillie Gifford Chair in the Ethics of Data and Artificial Intelligence at the Edinburgh Futures Institute (EFI) at the University of Edinburgh, where she is also appointed in Philosophy; she is also an affiliate scholar with Markkula Center for Applied Ethics. Views are her own.
First and most urgently, we need to consider what sources of data we already had and simply failed to use or act upon effectively; creating new surveillance data streams doesn’t help if the human and institutional failures that underlie the present pandemic remain unaddressed.
Second, we should carry out a sober analysis of where, if anywhere, any real data gaps are (in collection or analysis) that directly impede public health efforts and endanger global health, and what missing data will actually materially improve our ability to act & intervene.
The questions we should be asking, AFTER “why didn't the data we had prevent this?”, are “what further data do we need” and “from where/whom” and “how granular/sensitive?”—not just an indiscriminate grab for "moar data."
Furthermore, we should take the present crisis as a warning that even new and needed data are useless to gather if we can't restore public trust AND trustworthiness in the empowered bodies of expertise needed to apply that data for public good. If health experts and govt authorities gather even more sensitive data with potential for abuse, yet the public cannot or will not trust the vital new health info or guidance that supposedly justified the risk and privacy harms of its collection, then the effort is a net loss.
Moral of the story: what we most urgently need is rapid social analysis of the moral, political, and organizational failure points in our human systems that got us to this point with the data we have, and aggressive institutional reforms to remedy them.
What we do not need to do is to recklessly pour new streams of dangerously sensitive data into damaged or broken systems that have already proven incapable of using the public health data we have to prevent a pandemic.
Yes, we might need to develop some new tools and methods for collecting and rapidly analyzing public health data at global, national, and local scales, and for using data in ways that improve our ability to rapidly respond to emerging grave threats to human flourishing. But we cannot afford to keep ignoring or passively tolerating the moral failure points in our systems and civic institutions that are the true causes of the suffering to come.
Note: Vallor's comments were initially published as a series of tweets on Twitter.