Generative AI can act as a cognitive collaborator—boosting idea generation, synthesis and strategic framing—but may also blunt intrinsic motivation and evaluative rigor, creating a 'Creativity Paradox' that firms must design around.

Main Finding

The paper argues that Generative AI (GenAI) should be treated as a cognitive collaborator—structurally analogous to human creative cognition—rather than merely a productivity tool. By mapping procedural parallels between GenAI architectures and classical creativity theories, the authors propose a role-based integration framework (Creative Generators, Conceptual Synthesizers, Strategic Framers, Human-Centred Facilitators) and introduce the "Creativity Paradox": GenAI can boost output and exploration while simultaneously risking declines in intrinsic motivation, conceptual risk-taking, evaluative depth, authorship clarity, and epistemic integrity.

Key Points

• Conceptual reframing: GenAI mirrors human creative mechanisms (heuristic/constraint search, divergent ideation, associative recombination, iterative refinement), enabling it to co-orchestrate creative workflows without displacing human evaluative authority.
• Role-based framework: The paper distinguishes specific organisational roles that GenAI can play, aligning particular GenAI capabilities to distinct creative functions:
  • Creative Generators — breadth/fluency in ideation.
  • Conceptual Synthesizers — cross-domain recombination and brokerage.
  • Strategic Framers — scenario construction and contextual priming.
  • Human-Centred Facilitators — process orchestration, feedback loops, and accessibility.
• Creativity Paradox: While GenAI increases idea fluency and speed, it can (1) reduce intrinsic motivation and perceived authorship, (2) discourage high-risk/transformational exploration (conservatism toward normative outputs), and (3) weaken evaluative depth and accountability.
• Theoretical synthesis: Reinterprets foundational creativity models (Amabile’s componential model, Csikszentmihalyi’s systems model, Guilford’s divergent thinking, Simon’s heuristic search, Perry‑Smith & Mannucci’s network perspective, Anderson et al.’s process model) through GenAI’s procedural logic.
• Practical prescriptions: Advocate role-sensitive deployment, safeguards to protect intrinsic motivation and accountability, and governance for authorship and epistemic integrity.

Data & Methods

• Type: Theory-driven conceptual study and integrative literature synthesis.
• Methods used:
  • Review and reinterpretation of classical and contemporary creativity theories.
  • Mapping structural and procedural parallels between those theories and GenAI architectures (e.g., probabilistic generation, hierarchical models, feedback-driven refinement).
  • Development of a normative role-based integration framework grounded in the theoretical mapping and extant empirical insights from innovation, network, and organisational studies.
  • Extension of paradox theory to identify and conceptualise emergent tensions (the Creativity Paradox).
• No original empirical dataset or statistical analysis is reported; contributions are conceptual and prescriptive, intended to guide empirical testing and organisational practice.

Implications for AI Economics

• Productivity vs. Quality trade-offs: GenAI can raise measurable creative output (idea counts, speed-to-prototype) and lower marginal costs of ideation, but may reduce the probability of high-impact (transformational) innovations if intrinsic motivation and risk-taking fall—implying non-linear returns to GenAI adoption in R&D and creative sectors.
• Labour demand and task reallocation: Role specialization suggests complementarities rather than one-for-one substitution. Demand will shift toward tasks emphasizing evaluative judgment, strategic framing, domain expertise, and human facilitation—skills with higher wages/returns where human authority and accountability matter.
• Human capital and skill premiums: Skills tied to evaluation, synthesis, and risk-bearing may see rising returns; routine ideation and combinatorial search tasks may face downward wage pressure as GenAI handles volume and breadth.
• Organisational capital and absorptive capacity: Firms with better governance, integration strategies, and incentive systems that preserve intrinsic motivation are likely to capture more surplus from GenAI—raising returns to managerial capabilities and organisational design.
• Intellectual property, attribution, and market structure: Ambiguities over authorship and epistemic provenance can affect licensing, IP value, and contract design. Markets for creative outputs may reprice based on provenance, curation, and human endorsement signals.
• Incentives & innovation investment: If GenAI increases short-run output but depresses exploratory risk-taking, optimal R&D investment strategies and public policy (e.g., R&D tax credits, prizes, or grants) may need recalibration to sustain transformational innovation.
• Measurement and research agenda for AI economics:
  • Empirically quantify the Creativity Paradox: randomized experiments or firm-level panel studies measuring idea novelty, subsequent commercial success, worker motivation, and adoption intensity.
  • Estimate task-level substitution/complementarity elasticities between GenAI and human effort across ideation, synthesis, evaluation, and implementation.
  • Model dynamic effects on returns to human creative labor and firm-level productivity (e.g., endogenous choice of governance that mitigates motivational losses).
  • Study market implications of authorship and provenance rules on pricing, adoption, and welfare.
• Policy implications: Regulations and standards that clarify provenance, incentivize high-risk long-horizon research, and support retraining toward evaluative and facilitative roles can help align GenAI diffusion with social welfare and long-run innovation.

Suggested next empirical steps (for researchers/policymakers): - Design field experiments in firms that adopt role-based GenAI deployment to measure effects on idea novelty, implementation rates, worker motivation, and profitability. - Estimate how compensation schemes (intrinsic vs. extrinsic incentives) interact with GenAI use to affect creativity outcomes. - Analyze sectoral heterogeneity to identify where GenAI raises versus lowers the probability of breakthrough innovation.

Paper details: Choi, Park, Paydas Turan & Lee (2026). "Beyond the Creativity Paradox: A Theory-informed Framework for Role-based Integration of Generative AI in Organisational Creativity." Information Systems Frontiers. DOI: 10.1007/s10796-026-10746-y.