Managers who report stronger AI-enabled decision intelligence also report more resilient supply chains, driven by improved human–AI collaboration and enhanced dynamic capabilities. However, the finding rests on a single-country cross-sectional survey of perceptions, so causality and external validity remain unproven.

Main Finding

AI-enabled decision intelligence (AIDI) improves supply chain resilience both directly and indirectly. Specifically, AIDI increases human–AI collaboration and strengthens organizational dynamic capabilities (sensing, seizing, reconfiguring), and these mechanisms—individually and sequentially (human–AI collaboration → dynamic capabilities)—mediate the positive effect of AIDI on supply chain resilience. All hypothesized relationships were supported in survey data from 294 managers/professionals in Iranian organizations using PLS-SEM.

Key Points

• Concepts
  • AIDI: an organizational capability that embeds AI-generated analyses, predictions, and recommendations into managerial decision processes (decision intelligence), not just technology adoption.
  • Human–AI collaboration: systematic interaction where AI provides analytic scale and humans provide contextual interpretation, judgment, and ethical considerations.
  • Dynamic capabilities (DCT): sensing, seizing, and reconfiguring resources and routines under changing conditions.
  • Supply chain resilience: ability to anticipate, respond to, recover from, and adapt after disruptions.
• Theoretical framing
  • Dynamic capabilities theory explains how AI-generated intelligence must be integrated into managerial and organizational processes to produce resilience.
  • Complementarity: AI and human expertise are complementary; AI alone is insufficient — organizational processes determine value realization.
• Hypothesized relationships (tested and supported)
  • H1: AIDI → Human–AI collaboration (positive)
  • H2: AIDI → Dynamic capabilities (positive)
  • H3: Human–AI collaboration → Dynamic capabilities (positive)
  • H4: Dynamic capabilities → Supply chain resilience (positive)
  • H5: AIDI → Supply chain resilience (direct positive effect)
  • H6: Dynamic capabilities mediate AIDI → resilience; additionally, human–AI collaboration and dynamic capabilities sequentially mediate AIDI → resilience.
• Contributions
  • Recasts AI in supply chains as a decision capability (AIDI) rather than only a technological resource.
  • Integrates human–AI collaboration with dynamic capabilities to explain the mechanism from AI to resilience.
  • Empirically documents direct and mediated pathways linking AIDI to resilience.

Data & Methods

• Sample: survey of 294 managers and professionals involved in AI-supported supply chain decision-making in Iranian organizations.
• Analysis: partial least squares structural equation modeling (PLS-SEM) to test the proposed model and mediating pathways.
• Findings: All hypothesized direct and mediated relationships were statistically supported (including sequential mediation: AIDI → human–AI collaboration → dynamic capabilities → resilience).
• Notes/limitations (as reported or implied)
  • Cross-sectional survey design limits causal inference.
  • Sample context: firms/participants from Iran; generalizability to other institutional contexts should be tested.
  • Measurement details and control variables are not reported here; replication with longitudinal or objective outcome data would strengthen causal claims.

Implications for AI Economics

• Complementarity and returns to investment
  • Economic returns from AI in supply chains depend critically on complementary investments in human capital (training, decision processes) and organizational dynamic capabilities. Firms that invest only in AI technology may realize limited resilience or productivity gains.
  • AIDI increases the productivity of managerial skill and adaptive routines; thus AI adoption is likely skill‑biased within firms, increasing returns to complementary managerial and organizational capabilities.
• Risk and expected-loss economics
  • By strengthening sensing, seizing, and reconfiguration, AIDI can lower the expected costs of supply chain disruptions (reduced downtime, faster recovery), altering firms’ optimal risk-management choices (inventory, redundancy, insurance).
  • At sector level, widespread AIDI adoption with effective human–AI integration could reduce systemic vulnerability in supply networks, but heterogeneous adoption and capability gaps may create uneven resilience and new concentration/risk externalities.
• Strategy and firm heterogeneity
  • Firms should evaluate ROI for AI investments conditional on existing dynamic capabilities and human–AI collaboration readiness. There may be thresholds or complementarities (nonlinear returns) where AI pays off only after organizational changes.
  • Mergers, alliances, or restructuring that transfer organizational capabilities (not only tech) may amplify the value of AI assets.
• Labor-market and policy implications
  • Because human interpretation and collaboration remain central, policy and firm-level investments in upskilling managers and workers will be critical to translate AI into economic value.
  • Policymakers aiming to improve supply chain resilience should support both AI diffusion and complementary managerial/organizational capacity building (training programs, standards for human–AI decision workflows).
• Measurement and further research priorities for economists
  • Quantify the reduction in disruption costs attributable to AIDI (direct vs mediated effects) using firm-level longitudinal performance and disruption event data.
  • Model how complementarities between AI and organizational capabilities shape adoption thresholds, wage structure, and market structure.
  • Study general equilibrium and systemic effects: does widespread AIDI adoption lower aggregate volatility or create correlated vulnerabilities through common algorithms/models?
  • Evaluate heterogeneity across contexts (industry, country institutional quality) to assess external validity of the capability-mediated pathway.

Summary takeaway: AI delivers resilience value only when paired with human–AI collaboration and dynamic organizational capabilities. For economists and decision-makers, that implies evaluating AI investments in the context of complementary human capital and adaptive processes — otherwise expected economic gains from AI in supply chains may not materialize.