Platform Economics and Network Effects in the Age of AI

Platform Economics and Network Effects in the Age of AI is a primary research focus, examining how new artificial intelligence affects our current lives. Members from our center propose several different solutions to the problems arising from using AI.

Their academic articles about the subject are:

  • Michal Gal & Amit Zac, Is Generative AI the Algorithmic Consumer We are Waiting for?  (Network L. Rev, 2024)

The article provides the first controlled empirical test of Conversational Recommender Systems (CRS) influence on real purchasing decisions. CRS, utilizing large language models (LLMs) like GPT and Gemini, consistently increases consumer expenses. Customized GPT produced the highest average spending. These effects stem from subtle linguistic framing and increased exposure to premium brands, not differences in perceived product quality or trust. These findings position CRS as a potent choice architect with implications for regulatory oversight.

  • Raz Agranat & Michal Gal, Hub Power and Hub(uses): Power Dynamics in Platform Ecosystems (Antitrust L. J. 2025)

This article analyzes hub power, focusing on highly connected participants within platforms. Drawing on network science and microeconomics, it shows how hubs exert disproportionate control over interactions and value creation, affecting competition both within and between platforms. It develops a framework with four key determinants of hub power: attractiveness, platform dependence, switching feasibility, and countervailing power dynamics. The analysis demonstrates the wide-ranging antitrust implications for platform governance and legal remedies.

  • Raz Agranat & Michal Gal, The Microsoft Formula For Platform Power: Do Significant Network Effects Inevitably Generate Winner-Takes-All Dynamics? (University of Chicago – Law School and University of Haifa – Faculty of Law, 2025)

This article contests Microsoft’s assertion that markets with significant network effects inevitably result in a winner-takes-all outcome. It disputes the legal Formula that demonstrates network effects and high market share prices to having monopolistic power, arguing that this framework relies on outdated, simplifying assumptions. It explains that dramatic advances in network science now provide sophisticated tools to analyze platform dynamics. The goal is to propose an enhanced, modernized framework for evaluating platform market power.

  • Raz Agranat & Michal Gal, Fueling Concentration: Network Effects and AI Agents (Network L. Rev. 2025)

This article examines the AI agents’ market’s susceptibility to tipping driven by traditional network effects. As agents mediate commerce, they consolidate users’ bargaining power, creating a self-reinforcing feedback loop. Larger user bases yield augmented negotiation leverage, enabling agents to deliver better outcomes and attract more users. Dominant agents form powerful hubs in a scale-free market structure, outperforming rivals via size rather than technological superiority. The paper suggests targeting the preferential attachment mechanism to preserve competition.

  • Michal Gal & Daniel L. Rubinfeld, Algorithms, AI and Mergers,(Antitrust L. J. 2024)

This article focuses on merger control issues arising from the increasingly important role of algorithms based on artificial intelligence. It identifies six main functions of algorithms that may affect market dynamics, such as data collection and monitoring. The article demonstrates how such algorithms can exacerbate anti-competitive conduct in seven scenarios relating to unilateral and coordinated effects, including collusion and price discrimination. These findings are then translated into merger policy, showing how algorithms affect substantive and institutional features of merger control.

This article argues that explainability alone is insufficient for trustworthy AI, especially in complex or high-stakes areas. Instead, it proposes AI validation, ensuring reliability, consistency, and robustness of outputs, as a practical regulatory focus. Drawing from international policy comparisons, it offers a framework based on validation, third-party audits, harmonized standards, and liability reforms. This approach aims to better integrate powerful but opaque AI into critical domains while balancing innovation, accountability, and public trust.​

Back to Research