Agentic AI

John Horton

John Horton is an MIT Sloan economist who studies AI agents as autonomous economic actors, framing their value in terms of transaction cost reduction and improved decision quality in information-asymmetric markets.

title: "John Horton" type: person tags: [#autonomy, #tool-use, #human-ai-interaction] created: 2025-01-30 updated: 2025-01-30 status: stub

John Horton

MIT Sloan economist studying AI agents as economic actors, focusing on how autonomous AI systems reduce transaction costs and reshape market behavior.

Overview

John Horton is the Chrysler Associate Professor of Management and an associate professor of information technologies at the MIT Sloan School of Management. His research sits at the intersection of labor economics, market design, and information systems, with a particular interest in improving the efficiency and equity of matching markets. In recent work, Horton has turned his attention to Agentic AI as a new class of economic actor — autonomous software systems capable of strategic interaction, tool use, and economic transactions.

Horton's economic framing offers a distinctive lens: rather than focusing on technical architecture, he examines why people deploy AI agents and under what conditions they deliver value, emphasizing transaction cost reduction as the fundamental economic promise of the technology.

Contributions to Agentic AI

  • Co-authored research (with Peyman Shahidi and others) on the economic implications of AI-mediated transactions, defining AI agents as "autonomous software systems that perceive, reason, and act in digital environments to achieve goals on behalf of human principals, with capabilities for tool use, economic transactions, and strategic interaction."
  • Theoretical framework identifying two deployment scenarios for AI agents: (1) making higher-quality decisions than humans by overcoming information constraints and cognitive limitations, and (2) making decisions of similar or lower quality but at dramatically reduced cost and effort.
  • Research showing AI agents provide particular value in markets with information asymmetries (e.g., insurance, used cars, real estate) by continuously monitoring and cross-referencing data sources.
  • Analysis of AI agent value in high-counterparty, high-evaluation-cost markets such as startup funding, college admissions, and B2B procurement.

Affiliation

  • Institution: MIT Sloan School of Management
  • Title: Chrysler Associate Professor of Management; Associate Professor of Information Technologies

Key Works

Source Material

  1. Agentic AI, explained — MIT Sloan Ideas Made to Matter — Source for Horton's definitions, research framing, and quotes on transaction costs and information asymmetry.

Related Pages

See also: Agentic AI, Tool Use, Human-AI Collaboration, Sinan Aral, Kate Kellogg

Open Questions

  • How does the transaction-cost framing of AI agents connect to classical economics (Coase, Williamson) and what new theoretical extensions are required?
  • At what point do AI agents acting as economic counterparties raise antitrust or market manipulation concerns?
  • How does agent-to-agent negotiation change market equilibria compared to human-to-human or human-to-agent transactions?

Page type: person | Status: stub