AI in HR can backfire: instead of boosting strategic capability, AI-driven routines and metrics can fragment HR alignment and shift training toward short-term measurable tasks, risking long-run erosion of firm-specific skills and competitive advantage.

Main Finding

AI intended to improve alignment in human resource management can unintentionally cause alignment breakdown. Through four interrelated mechanisms, AI integration can weaken HR horizontal and vertical alignment, redirect skill development away from strategic needs, induce metric-driven employee behavior, and erode the workforce as a source of sustainable competitive advantage. The authors synthesize these dynamics into a four-state framework (Strategic Synergy, Contested Alignment, Strategic Fragmentation, Algorithmic Capture) that describes firm trajectories under AI-driven misalignment.

Key Points

• Theoretical framing: builds on strategic human resource management (SHRM) — horizontal alignment (coherence among HR practices) and vertical alignment (fit with overall strategy) — and the process-based model of HRM system strength.
• Four propositions (mechanisms) by which AI undermines alignment:
• AI operational requirements create logic conflicts across HR practices, weakening horizontal alignment.
• AI-driven routines cause employee skill development to drift away from firm strategic needs (skill-development drift).
• Employees default to behaviors that optimize observable metrics that AI emphasizes, rather than strategic or unmeasured goals (metric-aligned behavior).
• These dynamics collectively diminish the workforce’s value as a sustained competitive resource (loss of human-capital rents).
• Four organizational states characterize stages of AI-driven misalignment:
  • Strategic Synergy: AI complements and strengthens aligned HR and strategy.
  • Contested Alignment: Tensions emerge between AI requirements and existing HR logics.
  • Strategic Fragmentation: Divergence between employee skills/behaviors and strategy becomes widespread.
  • Algorithmic Capture: AI/metrics dominate decision-making and short-term measurable outcomes, eroding long-run strategic value.
• Contribution: a process-based theory identifying organizational risks of AI adoption in management systems and mapping potential trajectories of misalignment.

Data & Methods

• Methodological approach: conceptual/theory-building paper.
• Tools used: literature synthesis across SHRM and HRM system-strength models; development of four formalized propositions; integrative framework mapping dynamics into four organizational states.
• No primary empirical dataset or quantitative tests reported — the paper is explicitly theoretical and identifies mechanisms and testable hypotheses for future empirical work.
• Assumptions and scope: focuses on AI integration into HR/management systems and organizational processes; treats AI as an operational technology that instantiates particular metrics, routines, and decision logics.

Implications for AI Economics

• Firm-level returns to AI investment are endogenous to organizational alignment:
  • AI can reduce or reverse expected productivity gains if it induces misalignment between HR practices and strategy.
  • Cost–benefit analyses of AI should include alignment risks and downstream effects on human-capital value.
• Labor demand and skill composition:
  • AI may redirect on-the-job training and skills toward metricized tasks, accelerating skill-biased changes that favor measurable, routinized competencies over strategic, tacit skills.
  • Potential human-capital depreciation: long-run erosion of scarce strategic skills reduces firms’ ability to earn rents from workforce capabilities.
• Measurement and empirical research implications:
  • Heterogeneity in organizational state (the four states) implies heterogeneous treatment effects of AI in micro and macro studies; naive averages may misstate AI’s productivity impact.
  • Empirical models should control for organizational alignment, HR practices, and incentive/measurement regimes to identify causal effects of AI.
• Market structure and competition:
  • Algorithmic Capture can lead to short-run competition on observable metrics, potentially producing industry-wide declines in long-run innovation or quality if many firms optimize the same proxies.
  • Conversely, firms that maintain Strategic Synergy could secure competitive advantage, increasing returns to complementary investments (training, managerial practices).
• Policy and governance:
  • Policies promoting worker retraining, transparency of algorithmic objectives, and governance of performance metrics can mitigate misalignment risks.
  • Regulators and analysts should consider organizational complementarities and governance structures when assessing AI’s economic impacts.
• Modeling recommendations for economists:
  • Incorporate endogenous human-capital accumulation and alignment dynamics into models of AI adoption and productivity.
  • Model multi-period trade-offs between metricized short-run outputs and long-run strategic capabilities to capture path dependence and lock-in.
• Research agenda: test the four propositions empirically (panel firm data, HR practice surveys, natural experiments), quantify prevalence of organizational states, and measure how alignment mediates AI’s productivity and labor-market effects.