Industrial Policy for the Intelligence Age: Ideas to Keep People First

Summary

"Industrial Policy for the Intelligence Age: Ideas to Keep People First" is a 13-page policy document published by OpenAI in April 2026, with no individually named authors — issued in the company's institutional voice. It is framed explicitly as a conversation-starter, not a finished proposal: "intentionally early and exploratory," "a starting point for discussion," "not... a comprehensive or final set of recommendations." It focuses on the United States but states the conversation "must ultimately be global." It has two parts — "Building an Open Economy" (economic-sharing and labor-market proposals) and "Building a Resilient Society" (AI safety/governance) — of which the first carries the Georgist relevance.

Source-quality classification: this is an advocacy source — the self-published statement of a company that is itself a leading commercial developer of frontier AI and would be a direct object of any tax or wealth-fund policy it discusses. It is cited here only to represent OpenAI's own stated position, not as independent analysis.

The Core Argument / Findings

The document's economic proposals, in the order it presents them:

• Worker voice, "AI-first entrepreneurs," and "Right to AI." Formal channels for workers to shape AI deployment in their own workplaces; microgrants and shared back-office support for AI-enabled small businesses; and treating baseline AI access as a literacy-level public good with free or low-cost access points.
• "Modernize the tax base." AI may shift activity from labor income toward "corporate profits and capital gains," eroding the payroll-tax base funding Social Security, Medicaid, and SNAP. The remedy menu is explicitly plural and conditional: "increasing reliance on capital-based revenues — such as higher taxes on capital gains at the top, corporate income, or targeted measures on sustained AI-driven returns — and... exploring new approaches such as taxes related to automated labor," paired with "wage-linked incentives that encourage firms to retain, retrain, and invest in workers, similar to existing R&D-style credits." This is the closest the document comes to a "robot tax" — one hedged option among several, not a headline demand, and bundled with an incentive-preserving carrot.
• "Public Wealth Fund." The proposal closest to a sovereign wealth fund: "a Public Wealth Fund that provides every citizen — including those not invested in financial markets — with a stake in AI-driven economic growth." "Policymakers and AI companies should work together to determine how best to seed the Fund," which "could invest in diversified, long-term assets" and whose "returns... could be distributed directly to citizens." No funding formula, target size, or governance structure is specified — markedly less concrete than Norway's or Alaska's funds, both capitalized by statutorily defined resource-rent shares.
• "Efficiency dividends." Despite the name, not a government transfer: employers are urged to convert AI-driven cost savings into worker benefits (retirement matches, healthcare, four-day-workweek pilots, "benefits bonuses") — a firm-level mechanism, not rent capture.
• Adaptive safety nets, portable benefits, and care-economy pathways — automatic-trigger unemployment/wage insurance keyed to real-time labor metrics, benefits decoupled from a single employer, and a proposed caregiving "family benefit."
• Accelerated grid buildout via public-private partnerships to finance AI energy demand, with language about "sharing the upside with the public."

The document is candid about the risk it is responding to: "there is also a risk that the economic gains concentrate within a small number of firms like OpenAI... Workers using AI might well agree that it's increasing their productivity without believing they're seeing the benefits."

Relation to the Georgist Case

The Public Wealth Fund and the "returns... distributed directly to citizens" language are structurally analogous to the two canonical geoist rent-capture instruments — a sovereign wealth fund and a citizen's dividend — and the document's own framing (broad-based ownership stake, insulated from starting wealth) echoes the geoist argument for socializing a windfall rather than letting it accrue entirely to first movers. That resemblance is worth recording precisely because it did not come from Georgist advocacy: a frontier AI lab found the sovereign-wealth-fund/dividend structure to be the natural policy shape for an unearned, broadly generated surplus.

But the wiki's rent-gradient rule applies here at its steepest point. Land rent is the clean case: fixed supply, no incentive story to protect, a century of incidence evidence. Whether the surplus this document wants to redistribute is economic rent is contested and undecided — OpenAI's document never uses the word "rent" and never argues its profits, or AI profits generally, are unearned; it frames the issue as concentrated gains and "shared prosperity," a fairness/stability argument rather than a rent-theoretic one. Whether large AI-firm returns are genuine economic rent (the position argued by Korinek & Stiglitz's "innovator rents" analysis — see Korinek & Stiglitz — AI, Innovator Rents and Non-Distortionary Redistribution) or largely quasi-rent rewarding a real, risky bet (the Schumpeterian objection steelmanned at taxing quasi-rents kills innovation) remains open. Read this page as evidence a major AI firm itself judges some redistribution strategically or ethically necessary — not as evidence that AI profits are rents.

Nuances and Limits

• Advocacy, not analysis. OpenAI is proposing policy about a technology market it dominates commercially. It favors "public-private collaboration" over unilateral government action and cautions against "regulatory capture" and burdens on smaller companies — positions that also serve an incumbent's competitive interest. Read the tax and wealth-fund sections as OpenAI's stated position, not disinterested analysis.
• The proposals are conditional and undesigned. Every economic idea uses hedged language ("could," "should explore," "should work together to determine"). No tax rate, Fund size, funding source, or timeline is specified. This is a discussion opener, not draft legislation.
• Tax reform is explicitly paired with incentive-preserving carrots, consistent with — and, from an interested party, weak corroboration for — the Schumpeterian worry that untargeted taxation of innovation-era returns risks damaging the incentive it taxes.
• "Efficiency dividends" is not a public dividend — an employer benefits mechanism, distinct from the Fund's citizen-facing distributions, with no enforcement mechanism proposed.
• No disclosure of OpenAI's own tax posture or contribution. The document calls on governments and "AI companies" jointly to modernize tax policy without stating OpenAI's own effective tax rate or how it would help fund the wealth fund it proposes.

Bears On

• Problem: Corporate profits increasingly reflect economic rents — the admission that "economic gains concentrate within a small number of firms like OpenAI" and that workers may not see productivity gains reflected in pay is a rare admission-against-interest from an AI-profit-taking firm, worth citing as corroborating context, though it does not itself argue the profits are rent rather than reward for risk.
• Objection: Taxing quasi-rents kills innovation — OpenAI's insistence on pairing any capital/AI-return tax with wage-linked retention incentives, and its caution against "regulatory capture," lends real-world, if self-interested, corroboration to the objection's premise that actors closest to AI profits treat blunt taxation of those returns as a live incentive risk.
• Concept: Sovereign Wealth Fund and Citizen's Dividend — the Public Wealth Fund proposal is a structural cousin of both, evidence the instrument's basic shape has appeal outside the Georgist tradition, but flagged as undesigned and not resource-rent-funded, unlike Norway's or Alaska's funds.

See Also

• Sovereign Wealth Fund
• Citizen's Dividend
• Objection: Taxing Quasi-Rents Kills Innovation
• Corporate profits increasingly reflect economic rents
• Korinek & Stiglitz — AI, Innovator Rents and Non-Distortionary Redistribution
• Economic Rent

Sources

1. OpenAI, "Industrial Policy for the Intelligence Age: Ideas to Keep People First," April 2026. PDF — used for all proposal descriptions, direct quotations, and the document's own framing/caveats; fetched and read in full (all 13 pages) this session.