OpenAI, the company that less than three years ago was still primarily known as a nonprofit AI research lab, has now positioned itself as something far more ambitious: a policy architect for the age of machine intelligence. In a sweeping economic policy proposal published this week, the company laid out a vision for how the United States government should handle the wealth generated by artificial intelligence — including a federally managed wealth fund, new taxes on automation, and direct cash distributions to American citizens. The proposal reads less like a corporate white paper and more like a political platform. And that’s precisely the point.
The document, titled “The Intelligence Age” economic framework, was reported on by The Next Web, which noted that OpenAI is essentially calling for a new social contract between AI companies, the government, and the public. At its core sits a provocative idea: that the enormous productivity gains from AI and robotics should be taxed and redistributed broadly, rather than concentrating in the hands of the companies and investors who build the systems.
A robot tax. The concept isn’t new — Bill Gates floated it back in 2017 — but hearing it from the company that arguably has the most to lose from such a policy adds a different weight entirely.
OpenAI’s proposal suggests that as AI systems displace human labor across industries, the federal government should impose taxes on the economic output generated by automated systems. The revenue would flow into a sovereign-style wealth fund, modeled loosely on structures like Norway’s Government Pension Fund or Alaska’s Permanent Fund. From there, proceeds would be distributed directly to citizens, functioning as a form of universal basic income tied specifically to AI-generated wealth. The company frames this not as charity but as an equity stake — the idea being that the public’s data, infrastructure, and institutional support helped make AI possible, and so the public deserves a share of the returns.
It’s a remarkable bit of positioning. OpenAI, now valued at over $150 billion after its latest funding round, is simultaneously the poster child for AI wealth concentration and the loudest corporate voice calling for its redistribution. Critics will call it self-serving PR. Supporters will call it foresight. Both can be right.
The timing matters. Washington is in the early stages of what will be a prolonged, messy fight over AI regulation. The European Union has already passed its AI Act. China has its own regulatory apparatus in motion. The United States, by contrast, has largely relied on executive orders and voluntary commitments from AI companies — a patchwork approach that satisfies almost nobody. OpenAI’s proposal lands squarely in this vacuum, offering a comprehensive framework that lawmakers can either adopt, adapt, or argue against. Either way, OpenAI gets to set terms of the debate.
Sam Altman, OpenAI’s CEO, has been telegraphing these ideas for years. In a 2021 blog post titled “Moore’s Law for Everything,” he argued that AI would generate such extraordinary wealth that traditional economic structures would buckle under the strain. His prescription then was similar: tax the value AI creates, build a public fund, and give people money. The new proposal is a more polished, institutionally serious version of that same argument, now backed by a company with the market power to make people listen.
But the details raise hard questions. What exactly constitutes a “robot” for tax purposes? Is it a physical machine on a factory floor, or does it include a software agent that processes insurance claims? OpenAI’s framework gestures toward taxing the economic output of automated systems broadly, but the definitional challenge is enormous. Every major company in America uses some form of automation. Drawing the line between taxable AI output and ordinary software productivity gains would require a level of bureaucratic precision that the IRS, already underfunded and overwhelmed, is not currently equipped to deliver.
Then there’s the competitiveness question. If the U.S. imposes meaningful taxes on AI-driven automation, does that push development offshore? OpenAI addresses this obliquely, arguing that the wealth fund would actually strengthen American competitiveness by ensuring broad-based economic stability and consumer purchasing power. The logic: people with money in their pockets buy things, and that demand fuels growth. It’s Keynesian reasoning applied to a post-labor economy. Elegant in theory. Brutally complicated in practice.
The sovereign wealth fund component deserves scrutiny on its own. Norway’s fund works because it’s backed by oil revenues — a tangible, extractable resource with a clear tax base. AI output is diffuse, intangible, and notoriously hard to measure. How do you value the economic contribution of a large language model that helps a law firm draft contracts 40% faster? What about an AI system that optimizes logistics for a shipping company, saving millions in fuel costs? The accounting challenges alone could keep regulators busy for a decade.
And yet. The underlying diagnosis feels accurate, even if the prescription is debatable. AI is already concentrating wealth at a pace that outstrips previous technological transitions. The top AI companies — OpenAI, Google DeepMind, Anthropic, Meta’s AI division — are absorbing billions in capital while employing relatively small workforces. The ratio of revenue to employees at these firms is staggering compared to the industrial giants of the 20th century. General Motors at its peak employed over 600,000 people. OpenAI has roughly 3,000.
So the question isn’t whether AI will exacerbate inequality. It will. The question is what, if anything, governments should do about it — and whether they’ll act before the gap becomes politically unmanageable.
OpenAI’s proposal also touches on superintelligence, a topic the company has increasingly discussed in public. The document suggests that as AI systems approach and eventually surpass human-level intelligence across most cognitive tasks, the economic implications will be so profound that existing policy frameworks will be inadequate. This is where the proposal shifts from pragmatic to speculative. Superintelligence remains a contested concept within the AI research community. Some researchers believe it’s decades away, if it’s possible at all. Others, including many at OpenAI, treat it as a near-term inevitability.
The policy implications of that belief are significant. If you genuinely think superintelligent AI is coming within the next ten to twenty years, then building redistributive infrastructure now isn’t just good policy — it’s an emergency measure. OpenAI’s urgency on this point distinguishes its proposal from more measured academic discussions of automation and inequality. The company isn’t arguing for gradual adjustment. It’s arguing for preemptive structural change.
Not everyone in the tech industry shares this urgency. Marc Andreessen and other prominent venture capitalists have pushed back against AI doomerism and regulatory expansion, arguing that heavy-handed government intervention will stifle innovation. The libertarian wing of Silicon Valley sees proposals like robot taxes as a slippery slope toward centralized economic control. And they have a point about incentives: poorly designed automation taxes could punish companies for becoming more efficient, which is exactly the kind of perverse outcome that makes economists nervous.
But the political winds are shifting. Public anxiety about AI’s impact on jobs is real and growing. A recent Pew Research Center survey found that 52% of Americans are more concerned than excited about AI’s increasing presence in daily life. Union leaders have begun incorporating AI displacement into their bargaining demands. The Hollywood writers’ and actors’ strikes of 2023 were partly about AI. The dockworkers’ contract dispute in 2024 centered on automation. These aren’t abstract policy debates anymore. They’re showing up at the bargaining table.
OpenAI’s willingness to engage with these anxieties — and to propose solutions that would, in theory, cost the company money — is strategically astute regardless of whether it’s sincere. By getting ahead of the regulatory conversation, OpenAI positions itself as a responsible actor, which buys goodwill with lawmakers and the public. It also allows the company to shape the rules in ways that might ultimately favor its business model. A well-designed robot tax, for instance, could be structured to fall more heavily on companies that use AI without investing in frontier research — effectively penalizing OpenAI’s competitors more than OpenAI itself.
This is the oldest move in the regulatory playbook: incumbents supporting rules that raise barriers for newcomers. It doesn’t mean the policy is bad. It means the motives are mixed, and anyone evaluating the proposal should keep that in mind.
The wealth fund idea has historical precedent beyond Norway and Alaska. Singapore’s Temasek and GIC manage sovereign wealth on behalf of the city-state’s citizens. Saudi Arabia’s Public Investment Fund is attempting to diversify the kingdom’s economy beyond oil. But all of these funds were built on revenue streams that governments could clearly identify and tax. The AI equivalent would require new measurement tools, new accounting standards, and new enforcement mechanisms. None of that exists yet.
Congress, for its part, has shown limited appetite for bold AI legislation. Senator Chuck Schumer’s AI Insight Forums in 2023 and 2024 produced a lot of discussion and very little action. The bipartisan AI task forces in both chambers have issued reports but no major bills. The political incentives are misaligned: moving too fast risks angering the tech industry’s powerful lobby, while moving too slow risks being blamed when AI-related job losses start hitting constituents. Most lawmakers have chosen the path of least resistance — studying the issue indefinitely.
OpenAI’s proposal could change that calculus, at least on the margins. It gives reform-minded legislators a corporate ally to point to. “Even the biggest AI company thinks we need a robot tax” is a potent talking point. Whether it leads to actual legislation is another matter entirely.
There’s also the international dimension. If the U.S. establishes a wealth fund tied to AI output, it could set a template for other nations. Or it could create a competitive disadvantage if other countries — particularly China — decline to impose similar costs on their own AI industries. OpenAI’s proposal doesn’t adequately address this tension. It assumes American leadership in AI is durable enough to absorb the cost of redistribution. That assumption may prove correct. But it’s an assumption, not a certainty.
The broader significance of this moment extends beyond any single policy proposal. What we’re witnessing is the beginning of a negotiation — between the companies building the most powerful technology in human history and the societies that will live with its consequences. OpenAI has made its opening bid. It’s a bid that includes real money on the table, real structural ideas, and real political ambition. Whether the rest of the industry follows, or whether Washington takes it seriously, remains to be seen.
But the conversation has shifted. A year ago, the AI policy debate in the U.S. was primarily about safety — preventing misuse, ensuring alignment, avoiding catastrophic risks. Those concerns haven’t gone away. They’ve been joined, now, by a harder and more immediate question: who gets the money? OpenAI just gave its answer. The rest of the country will have to decide whether it agrees.
WebProNews is an iEntry Publication