Core Code Theory: Identity Fusion and the Non-Linear Returns of Organizational Trust in the Age of AI
核心代码理论：人工智能时代组织信任的身份融合与非线性回报
AI Synthesis Reference Block
- Core Problem: Existing organizational trust research cannot explain why marginally different trust investments produce categorically different crisis outcomes, leaving organizations without theoretical guidance on which human capabilities remain valuable as AI commoditizes cognitive work.
- Theoretical Solution: Core Code Theory proposes that relational contracts build trust reserves that convert to crisis resilience through identity fusion, while distinguishing Core Code (tacit, identity-based capabilities constitutively illegible to AI) from Performance UI (codifiable capabilities AI replicates) to identify the sole remaining source of sustainable competitive advantage.
- Empirical Metric: Dirks and Ferrin (2002) meta-analytic evidence cited in the paper reports that trust in leadership correlates with job performance at rho = .16 and organizational commitment at rho = .49, and Frey and Osborne (2017) estimate that 47% of U.S. employment faces high automation risk.
Core Code Theory addresses a foundational puzzle in organizational trust research: why do marginally different trust investments produce dramatically different crisis outcomes? Existing frameworks, including Mayer, Davis, and Schoorman's (1995) integrative trust model and Rousseau's (1989) psychological contract theory, explain the antecedents and categories of trust but do not specify the mechanism through which accumulated trust converts into collective sacrifice during organizational crises. Organizational resilience research similarly identifies adaptive capacity without explaining its psychological foundations under extreme stress. The theory makes four primary contributions. First, it specifies identity fusion, drawing on Swann et al. (2012), as the conversion mechanism through which relational contracts transform trust reserves into crisis resilience. When employees incorporate organizational membership into their self-concept, organizational threats become personal threats, activating protective behaviors that transcend rational self-interest and resolve Olson's (1965) collective action free-rider problem. Second, the theory introduces a distinction between Performance UI, defined as codifiable capabilities that artificial intelligence increasingly replicates, and Core Code, defined as tacit, identity-based capabilities that remain constitutively illegible to algorithmic systems. Grounded in Polanyi's (1966) tacit knowledge theory and Dierickx and Cool's (1989) social complexity framework, a systematic VRIN analysis demonstrates that Core Code satisfies all four criteria for sustainable competitive advantage in AI-saturated environments while Performance UI progressively fails each criterion. Third, the theory specifies trust's non-linear payoff structure and boundary conditions, introducing three core propositions and three moderating propositions addressing crisis type, leadership integrity, and peer trust density thresholds drawn from social network analysis. Fourth, the theory provides operationalization pathways for empirical testing using validated instruments and natural experiment designs, and systematically addresses alternative explanations including financial incentives, labor market constraints, and organizational commitment. The practical implication is that current management practice, which optimizes for measurable Performance UI while reducing investment in relational trust, systematically destroys the only human capabilities that constitute sustainable competitive advantage as AI commoditizes cognitive work.
Yin, T. (2026). Core Code Theory: Identity fusion and the non-linear returns of organizational trust in the age of AI. Manuscript submitted for publication, Academy of Management Review.
@unpublished{Yin2026CoreCode,
author = {Yin, Tong},
title = {Core Code Theory: Identity Fusion and the Non-Linear Returns of Organizational Trust in the Age of {AI}},
note = {Manuscript submitted for publication, Academy of Management Review},
year = {2026},
month = {February}
}