Joseph Stiglitz is a University Professor at Columbia University and the 2001 Nobel Laureate in Economics, awarded for work on information asymmetries and market signaling. He chairs Columbia's Institute of Politics, Economics and Global Affairs. In AI policy, he co-authored "Information in the Age of AI: Challenges and Solutions," the Digitalist Papers Vol. 2 essay arguing that information is a public good whose under-production is worsened by AI, supporting a regulatory response.
Background
Stiglitz served as Chief Economist and Senior Vice President of the World Bank (1997–2000) and as Chairman of the Council of Economic Advisers (1995–1997) under the Clinton administration. He was on the Stanford economics faculty from 1976 to 1988, then moved to Princeton, and subsequently to Columbia.
His prior scholarship grounds his AI-era arguments. The Economics of Information (with Sanford Grossman, 1980) argued that information asymmetries cause persistent market failures, work recognized in his Nobel award. Across Globalization and Its Discontents (2002), The Price of Inequality (2012), and The Road to Freedom (2024), he has argued that unregulated markets produce distributional failures the political system must address. He advocates a program he terms progressive capitalism, including stronger antitrust, public-option competitors to private monopolies, worker participation, and progressive taxation.
Positions on AI
Stiglitz's main AI-policy contribution is the essay "Information in the Age of AI: Challenges and Solutions," written with Màxim Ventura-Bolet for Digitalist Papers Vol. 2 and building on his work in information economics. Its core argument is that information is a public good: markets already under-produce it and over-produce mis/disinformation. The essay holds that AI sharpens three failures. First, an undersupply of good information, as AI intermediaries capture value from traditional producers without routing traffic or compensation. Second, an oversupply of mis/disinformation, summarized as "engagement through enragement," with lies cheaper to produce than verified truth. Third, an undersupply of correction, because fact-checking is itself a public good.
The essay identifies four AI shocks: improved production and dissemination of information, erosion of the producer business model, a falling relative cost of lies, and a "drone-war" verification-versus-deception arms race. Its central normative claim is that not all innovation is welfare-enhancing; innovations in exploiting market power are welfare-decreasing, and without appropriate regulation AI's effects on the information ecosystem may be net negative.
Within the Digitalist Papers, Stiglitz takes a pro-regulation, progressive position. He sits opposite Cochrane (libertarian deregulation) and closer to Bengio and Korinek/Lockwood, whose essays advance public-goods and redistribution framings.
Relationships
- supports: The Digitalist Papers (Stanford, Volumes 1–2) — his essay.
- related: AI Dividends (Universal Basic Capital, Digital Dividend, Global Dividend) / Bounty and Spread — adjacent concept pages.
- related: Algorithmic Pricing and Antitrust — Stiglitz's market-power framing extends here.
- related: Daron Acemoglu — fellow heterodox-progressive economist.
- contradicts: AI Snake Oil — Narayanan and Kapoor (2024) — Stiglitz takes AI's information-ecosystem effects seriously; Narayanan/Kapoor are more dismissive.
- related: AI and Productivity / The Simple Macroeconomics of AI — broader AI-economics literature.