Authors: Ana Brandusescu*, Centre of Interdisciplinary Research on Montreal, McGill University
Topics: Geographic Information Science and Systems
Keywords: artificial intelligence, AI, political economy, emerging technologies, policy
Session Type: Virtual Paper
Start / End Time: 11:10 AM / 12:25 PM
Room: Virtual 47
Presentation File: No File Uploaded
Mirroring other countries in the world, artificial intelligence (AI) is being substantially underwritten by federal and subnational levels of government. In Canada alone, investments of 2 billion dollars CAD have come from the federal government. One province, Quebec, has invested 1.3 billion and, in the City of Montreal, over 2 billion has been announced. Governments form part of a well-subsidized Canadian AI ecosystem, an interorganizational network of private investors, startups, public institutions, universities, and nonprofits.
Through semi-structured interviews of individuals associated with Canada’s AI ecosystem and a content analysis of public documents, I examine the geo-political economy of AI. I use Benson’s (1975) four dimensions of interorganizational equilibrium to examine: (i) domain consensus (e.g., shifting from a discourse of public accountability to a discourse of responsible AI); (ii) ideological consensus (e.g., government and industry coalescing around a vision of economic growth that emphasizes AI); (iii) positive evaluation (e.g., actors across three provinces with central nodes in AI research nonprofits participating in a revolving door of personnel); and (iv) work coordination (e.g., collaboration between AI research central nodes, delegation to public-private leadership, co-creation of procurement processes for AI services).
Examined through the lens of interorganizational equilibrium, results reveal the lack of regulatory mechanisms to keep industry accountable to the public, the emergence of geographic arenas of conflicts, the prevalence of non-disclosure agreements for civil society, lobbying for industry funding that differs from traditional approaches, a revolving door for AI staff, and the formation of a kind of AI sovereignty.