AI Research Intelligence Briefing

Institutional Theory (DiMaggio and Powell, 1983). Introduction: Institutional theory examines how organizations respond to pressures from their social environment. Organizations in the same field become isomorphic through coercive, normative, and mimetic pressures. Legitimacy is attained by conforming to institutionalized rules, norms, and cognitive expectations. Firms that deviate face penalties in resources, reputation, and survival. Fit: As safety advocacy grows visible, AI labs face normative pressure from researchers and coercive pressure from regulators to adopt visible safety governance. Altman’s deceleration signal is consistent with a legitimation response to these combined pressures. References: DiMaggio, P. J., and Powell, W. W. (1983). American Sociological Review, 48(2), 147–160. Scott, W. R. (1995). Institutions and Organizations. Sage.

Commons Governance (Ostrom, 1990). Introduction: Elinor Ostrom’s framework examines how communities manage shared resources to avoid the tragedy of the commons. Effective governance requires boundary rules, monitoring, graduated sanctions, and collective-choice arrangements. Fit: Frontier AI capabilities represent a common resource whose over-exploitation via competitive racing may generate societal harm. Altman’s deceleration proposal is consistent with a polycentric governance attempt, requiring multi-lab coordination rather than unilateral restraint. References: Ostrom, E. (1990). Governing the Commons. Cambridge University Press.

Agency Theory (Jensen and Meckling, 1976). Introduction: Agency theory models the principal-agent relationship in which principals delegate tasks to agents who may have divergent interests. Information asymmetry creates risks of moral hazard and adverse selection. Monitoring, incentive alignment, and bonding reduce agency costs. Fit: AI developers may advance capabilities faster or in directions diverging from the interests of users and the public. Safety researcher whistleblowing, voluntary deceleration, and IVO auditing all represent mechanisms to reduce principal-agent information asymmetry. References: Jensen, M. C., and Meckling, W. H. (1976). Journal of Financial Economics, 3(4), 305–360.

Platform Ecosystem Theory (Gawer and Cusumano, 2002; Parker et al., 2016). Introduction: Platform ecosystem theory examines how platform owners attract and govern complementors and users to generate network effects and competitive advantages. Platforms compete on ecosystem breadth and quality. Multi-homing and switching costs determine lock-in. Fit: The Moonshot-Kimi challenge illustrates how low-cost entrants with competitive capabilities can disrupt platform lock-in, particularly when switching costs for API-based services are low and enterprises multi-home across providers. References: Gawer, A., and Cusumano, M. A. (2002). Platform Leadership. Harvard Business School Press. Parker, G., Van Alstyne, M., and Choudary, S. (2016). Platform Revolution. Norton.

Resource-Based View (Barney, 1991). Introduction: The RBV holds that firms achieve competitive advantage through resources that are valuable, rare, inimitable, and non-substitutable. Dynamic capabilities extend this to the capacity to reconfigure resources in changing environments. Fit: Moonshot’s cost-leadership strategy challenges whether US AI labs hold VRIN advantages through model quality alone. If inference cost is a primary competitive dimension, cost-efficient infrastructure becomes the key resource, which is more imitable. References: Barney, J. (1991). Journal of Management, 17(1), 99–120.

Sociotechnical Systems Theory (Trist and Bamforth, 1951; Orlikowski, 1992). Introduction: Sociotechnical systems theory holds that technological change and social/organizational change must be jointly optimized to realize productivity gains. Technology alone does not determine outcomes; how work is reorganized around technology matters critically. Fit: The UK GDP evidence suggests AI has begun to manifest in macroeconomic output, but whether productivity gains depend on sociotechnical reconfiguration versus pure tool substitution remains empirically open, directly testable using sector and firm data. References: Trist, E. L., and Bamforth, K. W. (1951). Human Relations, 4(1), 3–38. Orlikowski, W. J. (1992). Organization Science, 3(3), 398–427.

Task-Technology Fit (Goodhue and Thompson, 1995). Introduction: Task-technology fit theory proposes that technology produces performance gains when it matches the requirements of the tasks it is used to perform. The theory generates predictions about which task types benefit most from a given technology. Fit: The UK data shows computer programming benefiting most from AI, consistent with high task-technology fit for coding and data analysis. IS research can test which specific task types drive macro productivity gains using occupational data cross-referenced with AI tool usage rates. References: Goodhue, D. L., and Thompson, R. L. (1995). MIS Quarterly, 19(2), 213–236.

Institutional Theory (DiMaggio and Powell, 1983; Meyer and Rowan, 1977). Introduction: Institutional theory explains how organizations conform to the expectations of their social environments to gain legitimacy. Coercive isomorphism occurs when organizations comply with legal requirements. Normative isomorphism occurs when professional norms shape practices. Mimetic isomorphism occurs when firms imitate leaders under uncertainty. Fit: The IVO framework introduces coercive institutional pressure on AI developers. The spread of AI governance laws across 27 states indicates emerging field-level norms that may produce mimetic adoption of IVO-type practices even by non-California firms. References: Meyer, J. W., and Rowan, B. (1977). American Journal of Sociology, 83(2), 340–363. DiMaggio and Powell (1983), American Sociological Review, 48(2), 147–160.

Control Theory (Ouchi, 1979; Kirsch, 1997). Introduction: Control theory examines formal and informal mechanisms through which organizations and regulators manage agent behavior. Outcome control monitors results; behavior control monitors processes; clan control relies on shared values. Portfolio governance uses combinations of these modes. Fit: The IVO auditing framework represents a new form of behavior control applied to AI systems by regulatory mandate. The prohibition on neural data collection represents an absolute boundary-setting form of control. IS research can examine how different control modes affect AI deployer behavior differently. References: Ouchi, W. G. (1979). Management Science, 25(9), 833–848. Kirsch, L. J. (1997). Information Systems Research, 8(3), 215–239.

Attention-Based View of the Firm (Ocasio, 1997). Introduction: The attention-based view proposes that firm behavior reflects the allocation of organizational attention to different issues by decision makers. What executives attend to determines what organizations do. Structural factors channel attention, and attention in turn shapes strategic choices. Fit: When the president frames AI development as a China-competition issue, he allocates national-level attention to competitiveness rather than safety. AI firms face attention-channeling effects that may redirect internal resources away from safety investment. References: Ocasio, W. (1997). Strategic Management Journal, 18(S1), 187–206.

Institutional Logics (Friedland and Alford, 1991; Thornton et al., 2012). Introduction: Institutional logics are overarching belief systems and material practices of a societal order that guide organizational and individual action. When multiple logics compete, organizations must navigate institutional complexity through decoupling, compromise, or selection. Fit: Trump’s safety dismissal reveals a direct clash between a geopolitical-competition logic and a safety logic. AI firms and regulators must navigate this institutional complexity, and their navigation strategies are a direct IS research object. References: Friedland, R., and Alford, R. R. (1991). In W. W. Powell and P. J. DiMaggio (Eds.), The New Institutionalism in Organizational Analysis. University of Chicago Press. Thornton, P. H., Ocasio, W., and Lounsbury, M. (2012). The Institutional Logics Perspective. Oxford University Press.