Entrepreneurs do more than exploit property rules — they make them: a new typology shows how firms actively create product, asset and resource rights, reshaping markets, firm boundaries and rents; applied to AI, these strategies determine who controls models, data and scarce compute, with large implications for competition and welfare.

Main Finding

Entrepreneurs are not merely passive users of existing legal and economic institutions; they play a central, active role in creating and reshaping property rights. The paper develops a typology of "property-rights entrepreneurship" — product rights, asset rights, and resource rights — and analyzes the entrepreneurial challenges, actions, and outcomes associated with each. It argues for treating property-rights formation as an entrepreneurial activity and calls for further research at the intersection of managerial economics, entrepreneurship, and property-rights theory.

Key Points

• Property rights are foundational to market exchange because they determine excludability and thus the incentives for entrepreneurial activity.
• Entrepreneurial activity itself is a driver of property-rights evolution; entrepreneurs invent, claim, contest, and institutionalize new forms of excludability.
• The authors distinguish three types of property-rights entrepreneurship:
  • Product rights: rights that govern exclusion over particular product outputs (e.g., proprietary software features, product-level IP, licensing of end-products).
  • Asset rights: rights over specific assets that firms hold (e.g., ownership and control of datasets, trained models, specialized equipment).
  • Resource rights: rights over scarce or rival inputs/resources (e.g., access to compute, spectrum, exclusive data sources, platform access).
• For each type the paper outlines typical entrepreneurial challenges (e.g., defining boundaries of exclusion, enforcement, coordination with institutions), common actions (e.g., legal claims, technological design, contractual arrangements, standard setting), and potential outcomes (new markets, shifts in firm boundaries, reallocation of rents, institutional change).
• The approach is conceptual and synthetic: it reframes property-rights theory by placing entrepreneurs and managerial strategies at the center of how property regimes emerge and change.

Data & Methods

• From the abstract, the paper appears to be primarily conceptual/theoretical and typological: a literature synthesis that builds a framework distinguishing three modes of property-rights entrepreneurship.
• Methods likely include:
  • Review and integration of literatures from property-rights economics, entrepreneurship, and managerial economics.
  • Development of a conceptual typology with illustrative examples (rather than new large-scale empirical datasets).
  • Identification of research gaps and propositions for future empirical work.
• Opportunities suggested (explicitly or implicitly) for empirical approaches include case studies, historical institutional analysis, comparative legal-institutional studies, field experiments on exclusion/enforcement mechanisms, and econometric analysis of firm behavior around claims over products, assets, and resources.

Implications for AI Economics

• A useful lens for AI: AI commercialization involves intense property-rights entrepreneurship across all three types:
  • Product rights in AI — e.g., firms crafting licensing regimes and API constraints around model outputs, selective feature gating, or claim over generative outputs.
  • Asset rights in AI — ownership and control of datasets and trained models (weights), which are essential firm assets that determine competitive advantage.
  • Resource rights in AI — exclusive or prioritized access to scarce compute, curated data sources, labeling capacity, or talent.
• Specific implications and predictions:
  • Entrepreneurial strategies (open-source vs. proprietary, licensing, platform access controls) will shape how excludable AI outputs are and therefore the market structure of AI industries.
  • Innovations in contractual and technological design (fine-grained licensing, watermarking, secure enclaves, differential privacy, model card standards) are mechanisms entrepreneurs use to create or enforce new property boundaries.
  • New forms of assetization (tokenization of model access, data trusts, model-as-a-service subscriptions) will reconfigure investment incentives, firm boundaries, and the allocation of rents between startups, platforms, and users.
  • Competition for resource rights (e.g., privileged compute capacity, exclusive datasets) can generate winner-take-most dynamics, raise entry barriers, and influence industrial concentration in AI.
  • Regulatory and institutional responses (intellectual property law, data-protection regimes, antitrust enforcement) will interact with entrepreneurial actions, shaping whether emergent property regimes are private, public, or hybrid.
• Directions for AI-economics research inspired by this framework:
  • Empirically measure how AI startups' choices (open-source vs. closed-source; API gating) affect market entry, diffusion, and innovation rates.
  • Study how novel contractual forms and standards (e.g., licenses for models/datasets) emerge and stabilize — and who benefits.
  • Analyze the effect of exclusive resource control (compute contracts, data partnerships) on competition and welfare in AI ecosystems.
  • Examine legal and technological enforcement effectiveness (watermarks, provenance) and their economic consequences for excludability and secondary markets.
  • Investigate institutional entrepreneurship in AI governance: how firms, consortia, and regulators co-construct property regimes for AI artifacts.
• Overall: framing AI-related ownership questions as instances of property-rights entrepreneurship highlights entrepreneurs as causal agents shaping the incentives, distribution, and pace of AI innovation — and points to concrete empirical and policy-relevant research agendas.