Understanding the Debates on Facial Recognition Technology: A Network Analysis

• In Event: New Challenges in Technology Governance

Abstract

The widespread use of facial-recognition technology (FRT) presents major human rights challenges; yet outside of the European Union (EU), few legally binding rules apply to FRTs. To address this puzzle, we used complex systems theory to analyze political debates on FRT. We argue that these debates are embedded in networks of actors and ideas that influence policy-making processes. We used a mixed-method research design combining network analysis and structured interviews of key actors in order to trace the evolution of political debates and better understand the role of different actor coalitions in major policy decisions.
First, we used Discourse Network Analysis (DNA) to trace the evolution of political frames in the United States, the United Kingdom, and the EU. Our analysis of 4000 political statements between 2000 and 2022 show that FRT has been framed as a solution for security and efficiency issues, particularly in the aftermath of the 9/11 attacks. However, since 2015, these frames have been increasingly challenged, shifting the discourse to concerns about lack of regulation, unlawful surveillance, bias, and privacy. The growing salience of critical frames and non-state actors in the EU have resulted in demands for tighter regulation of FRT, including a push for laws targeting the commercial exploitation and potential abuse of human data. Second, we used the network visualization and the quantitative data produced by the DNA to identify 35 actors that played central roles in their respective clusters. We conducted structured interviews with these actors to further substantiate their roles in shaping the debate and relevant policies.
Our comparative network analysis and its evolution over successive time slices help us better understand the policy asymmetry between the Europe and the United States with regards to facial recognition. Our argument is centered around the interchanging role that non-governmental organisations (NGOs) and technology firms play in the US and European networks. While both sets of actors are precursors in the debate around FRT and they influence supranational organisation and national governments, their roles are different in the European and American networks. In Europe, NGOs play a central role in structuring the network which explains the evolution of the debate toward the topics of regulation and ethical concern. In the United States, technology firms play a more dominant role, which explains the lack of federal regulation.
The qualitative data drawn from interviews shed further light into the regulation and policy gap between Europe and the United States on facial recognition. The US actors are more positive towards the development of FRT and favour minimal regulation to address its implementation by security forces. Meanwhile European actors tend to favor stronger restrictions and focus mostly on human rights and privacy concerns as the main reasons to regulate the technology. Overall, our results show a significant link between actor and frame configurations within the FRT network and policy outcomes that impact citizens’ democratic rights and privacy.

Authors

• Manuel Quintin, Université Laval

• Kerem Öge, University of Warwick