OpenAI 提議徵收機械人税、設立公共財富基金及試行四日工作週

OpenAI has published a 13-page policy document calling on the U.S. government to tax automated labor, create a national wealth fund seeded by AI companies, and pilot 32-hour work weeks at full pay – a package of proposals the company frames as essential preparation for superintelligent AI.

The document, titled “Industrial Policy for the Intelligence Age: Ideas to Keep People First,” marks a sharp escalation from OpenAI’s January 2025 Economic Blueprint, which focused on infrastructure investment and light-touch regulation. The new paper directly addresses wealth redistribution, job displacement, and the limits of existing social safety nets.

OpenAI CEO Sam Altman said the company feels urgency about starting these conversations now rather than after displacement begins. The full document is framed as a starting point for debate, not a final set of recommendations.

What the Document Proposes

The six core proposals span fiscal policy, labor markets, and AI containment.

First, a Public Wealth Fund would give every American citizen a direct stake in AI-driven economic growth. The fund would be nationally managed, seeded in part by AI companies, and would invest in diversified, long-term assets across both AI firms and businesses deploying the technology. This is the most structurally ambitious item – effectively proposing a sovereign wealth fund tied to AI profits.

Second, the document calls for taxes on automated labor and a shift in the tax base from payroll toward capital gains and corporate income. OpenAI’s reasoning: if AI displaces enough workers, the wage-and-payroll tax revenue that funds Social Security, Medicaid, and SNAP will collapse. Taxing the capital that replaces labor is the proposed fix.

Third, OpenAI suggests the government incentivize four-day work weeks at full pay, framing the reduced hours as an “efficiency dividend” – converting AI-driven productivity gains into time back for workers rather than purely into corporate margins. The company recommends pilots involving both employers and unions.

Fourth, the plan frames AI access as a right, comparable to literacy, electricity, and internet access. It calls for affordable access for workers, small businesses, schools, libraries, and underserved communities – a notable position from a company whose commercial expansion and financial integrations have centered premium pricing.

Fifth, OpenAI includes containment playbooks for scenarios where dangerous AI systems cannot be easily recalled because they are autonomous and capable of self-replication. The document proposes government coordination as the mechanism for managing these risks.

Sixth, the document proposes auto-triggering safety nets tied to economic data. When AI displacement metrics hit preset thresholds, temporary increases in unemployment benefits, wage insurance, and cash assistance would kick in automatically. When conditions stabilize, the measures would phase out.

Why Now, and Why From OpenAI

The timing is strategic. OpenAI closed a $110 billion funding round earlier this year at a valuation exceeding $800 billion, making it the most valuable private technology company in history. Its models now reach over 100 million weekly users in India alone. The scale of deployment gives weight – and self-interest – to its policy positions.

Altman framed the current moment as comparable to the Progressive Era of the early 1900s and the New Deal of the 1930s, periods when American institutions were rebuilt around new economic realities. He also flagged immediate threats from AI-enabled cyberattacks and biological weapons as near-term risks that require coordinated government response.

OpenAI is also establishing a pilot program of fellowships and research grants of up to $100,000, plus up to $1 million in API credits, for researchers building on the document’s policy ideas. The company plans to open an OpenAI Workshop in Washington, D.C. in May to host ongoing policy discussions.

The document positions OpenAI in territory Anthropic has occupied with its emphasis on responsible development. Other government AI workforce programs have already started appearing internationally, suggesting the policy window OpenAI is targeting is real.

Whether these proposals survive contact with Congress or remain a corporate positioning exercise is an open question. But a company valued at $800 billion publicly calling for taxes on the technology it sells is, at minimum, a signal that the industry’s own leaders see disruption as imminent rather than theoretical.

Alex leads Unite.AI’s AI-powered news operations, combining journalism, research, and automation to support timely and scalable coverage of artificial intelligence. His work helps ensure emerging AI developments are surfaced efficiently while maintaining the publication’s editorial standards.