Platforms that visibly innovate and promise to share critical resources build reputations that raise partners’ willingness to trust and participate; however, perceived opportunism undermines that conversion while dominant platforms with strong manipulation power amplify the effect of reputation.

Main Finding

Core firms’ ecological niche creativity and willingness to share key resources increase subordinate actors’ intention to trust those core firms; this effect operates through the core firms’ reputation. Opportunistic behaviour by core firms weakens the positive link between reputation and actors’ trust, while the core firms’ manipulation power strengthens it.

Key Points

• The study adopts signalling theory to explain how core firms communicate qualities that shape trust among ecosystem actors.
• Two positive signals from core firms:
  • Ecological niche creativity (innovation in the platform ecosystem).
  • Key resource sharing intention (willingness to share data, APIs, tools, or other critical inputs).
• Reputation of the core firm mediates both relationships: signals → reputation → actors’ trust intention.
• Two contextual moderators:
  • Opportunistic behaviour (self-serving actions by the core firm) negatively moderates reputation → trust: reputation converts less effectively into trust when opportunism is perceived.
  • Manipulation power (ability of the core firm to shape outcomes or control ecosystem dynamics) positively moderates reputation → trust: reputation matters more for trust when the core firm has high manipulation power.
• The study foregrounds actor-centric, social mechanisms of trust in platform ecosystems, contrasting with prior research emphasizing algorithmic or technical trust mechanisms.

Data & Methods

• Empirical base: survey of 347 subordinate actors operating within digital platform ecosystems.
• Conceptual framework: signalling theory with a mediated pathway (signals → reputation → trust intention) and moderated effects on the reputation → trust link.
• Analytical approach: mediation and moderation tests (likely structural equation modeling or regression-based mediation/moderation techniques).
• Measures: constructs such as ecological niche creativity, key resource sharing intention, reputation, opportunistic behaviour, manipulation power, and trust intention were measured via actor self-reports (presumably Likert scales).
• Limitations (inferred): cross-sectional survey design limits causal claims; sample drawn from actors in unspecified platform types may limit generalizability; reliance on perceptual measures.

Implications for AI Economics

• Participation and resource flows
  • Trust-building signals (innovation orientation and resource-sharing commitments) encourage firms to join and contribute to AI platform ecosystems, increasing shared data, models, and tooling that fuel AI development and network effects.
• Incentive and governance design
  • Platforms should cultivate reputational signals (transparent roadmaps, shared governance, visible collaboration histories) and design incentives to reduce opportunism (contracts, SLAs, auditability, dispute resolution) to sustain cooperation.
• Market power and welfare
  • Manipulation power amplifying the role of reputation implies that dominant platforms can convert reputation into stronger control over ecosystem behavior—raising concerns about lock-in, rent extraction, and competitive distortions in AI markets. Antitrust and regulatory attention should consider how reputational mechanisms interact with market power.
• Data- and model-sharing policies
  • Encouraging resource sharing (e.g., safe APIs, data trusts, standardized interfaces) can enhance trust and accelerate collective AI innovation, but must be paired with safeguards against opportunistic misuse.
• Measurement and governance of trust beyond algorithms
  • Economic models of platform competition and cooperation should incorporate reputational signalling and strategic opportunism as determinants of participation, not just algorithmic characteristics or pricing.
• Research and policy priorities
  • Future work should test causal pathways (e.g., longitudinal or experimental designs), disaggregate effects by platform type (marketplaces vs. developer platforms vs. data ecosystems), and examine how algorithmic governance (e.g., automated moderation, recommender transparency) interacts with interpersonal and reputation-based trust.