Redrawing data boundaries: From private collection to public good
Can data collected by private companies be recognized as a public resource to be leveraged for social good? How can we recognize data as a public resource while still protecting fundamental individual rights? In a panel discussion at Absolutely Interdisciplinary 2022, SRI Associate Director Lisa Austin discussed the shape of data today with Eric Horvitz (Microsoft), Aziz Z. Huq (University of Chicago), Robert Seamans (Stern School of Business, NYU), and Pamela Snively (TELUS).
The immense amounts of data collected and processed by companies through technological tools are fenced by boundaries that can be legally constructed through various means, including intellectual property rights, privacy rights, and contracts. Powering everything from strategic decision-making to advanced technologies like artificial intelligence systems, data is unquestionably one of the most valuable resources for our society today.
As data continues to shape and redefine the world in new ways, how might we redraw new boundaries for the use of data, where it can be recognized as a public resource? Can we accomplish this while also ensuring that the collection of data about persons respects the privacy interests of the subjects affected? What role can private companies play in making data available for greater public access to researchers, governments, and civil society—and, in doing so, help leverage it for social good?
At the Schwartz Reisman Institute for Technology and Society’s (SRI) recent conference, Absolutely Interdisciplinary 2022, SRI Associate Director and University of Toronto Professor of Law Lisa Austin posed these questions in a session entitled “Redrawing Data Boundaries.” The session’s panelists comprised several leading thinkers on the role of data in society: Aziz Z. Huq, a professor of law and scholar of US and comparative constitutional law at the University of Chicago; Robert Seamans, an associate professor at New York University’s Stern School of Business, and former Senior Economist for Technology and Innovation on President Obama's Council of Economic Advisers; Pamela Snively, Chief Data and Trust Officer at TELUS; and Eric Horvitz, Chief Scientific Officer at Microsoft, and an advisor on the US President’s Council of Advisors on Science and Technology.
Panelists in “Redrawing Data Boundaries” at Absolutely Interdisciplinary 2022 (from left to right): SRI Associate Director Lisa Austin, Eric Horvitz (Microsoft), Aziz Z. Huq (University of Chicago), Robert Seamans (Stern School of Business, NYU), and Pamela Snively (TELUS).
What are the benefits and risks of greater public access?
Increasing public access to data collected by private companies can offer a wide range of social benefits. For example, Snively discussed TELUS’ Data for Good program and how sharing data responsibly can have an impact in helping governments make evidence-based public policy decisions. During the COVID-19 pandemic, the Data for Good program provided research partners with de-identified network mobility data to help with pandemic response initiatives.
However, data sharing can also raise concerns about trust and reputational risk for companies who are willing to engage in such initiatives. As Snively explained, “Sharing outside the organization, we end up in a situation where consumers must trust the company, the government, and the rules and commitments set by the government. Where trust can be compromised by allowing others to access data, there is a risk of a chilling effect on consumer participation in the digital economy.”
Additionally, while Horvitz offered support for data sharing initiatives, he also cautioned the need to carefully consider risks inherent in government access when data sharing may not be aligned with democratic values or could negatively affect privacy and civil liberties. Horvitz also highlighted the existence of legal gaps, such as large-scale commercial datasets collected by third-party companies that can be purchased by governments without warrants and used for multiple purposes, including building predictive models.
How we conceive of data defines how we use it
Panelists also considered how the potential redrawing of data boundaries requires careful consideration of privacy concerns with regards to the ownership of personal data, and issues of consent and control. As Huq highlighted in his remarks, under current US law it’s unclear whether personal data produced by tech companies fall under personal ownership or ownership by a corporate entity. Huq described how personal data is no longer just personal, but inexorably social—for example, data collected from one group of people can in turn be used to make predictions about others.
Horvitz underscored that requiring tech companies to share data would require new kinds of data agreements that allow individuals more control over their personal data, and the ability to consent to how their data may be used or shared. However, Huq cautioned that exercising individual control over one’s data has become more challenging in recent years, as it becomes more difficult to predict how one’s data could be used. In contrast, a legal foundation for regulatory intervention may better help re-orient data usage for the public good.
In Canada, the proposed law Bill C-27, called the Digital Charter Implementation Act 2022, aims to increase personal control over one’s information and how digital platforms handle personal data. This bill requires data mobility frameworks to allow for data portability—the ability for individuals to transfer data collected about them by a particular company to be ported to another. Seamans suggested data portability might act as a middle ground between two competing conceptualizations of data, as an asset fenced by private companies versus a publicly available resource.
Seamans described how data portability would impact not just big tech, but a wide range of other industries, including healthcare and banking: “My bank has valuable data. They know exactly how much I make. They know what I spend it on. They know the extent to which I save it or invest it. That is valuable data that I might want to bring to a start-up, credit union, or local competitor.”
As the discussion demonstrated, responsible data sharing offers significant potentials for social benefits, but its successful implementation will require updating how we conceptualize our personal data and ownership, building public trust, and using regulatory frameworks to orient data sharing for social good. As the panel concluded, participants agreed on the need for greater data literacy and education for the public at large, and that the ways we choose to conceive of data have significant impacts on how we value and use it as a society.
Watch the recording:
About the author
Lief Pagalan is a PhD student in Epidemiology at the Dalla Lana School of Public Health at the University of Toronto, and a 2021–22 Schwartz Reisman Graduate Fellow. Their research develops new applications for machine learning in public health to improve the built environment and population health. Pagalan holds an MSc in Health Sciences and a BSc in Geography from Simon Fraser University, and a BFA from OCAD University.
