As governments grapple with how to manage the economic fallout of superintelligent machines, OpenAI has released a series of policy proposals outlining how wealth and work could be reshaped in the “age of intelligence.” The idea blends traditionally left-leaning mechanisms, such as public wealth funds and expanded social safety nets, with a fundamentally capitalist, market-driven economic system.
OpenAI’s proposal is essentially a wish list and public manifesto to help elected officials, investors and the public understand how the $852 billion company sees the world changing in an era where artificial intelligence is transforming work and the economy.
This proposal was announced amid increasing anxiety surrounding AI due to concerns about job replacement, concentration of wealth, and expansion of data centers across the country. They also arrive as the Trump administration moves toward a national AI framework and heralds an attempt at bipartisan positioning ahead of the midterm elections. These efforts come alongside a more direct political push. OpenAI Chairman Greg Brockman, who donated millions to President Donald Trump, and other tech billionaires have invested hundreds of millions of dollars in super PACs that support light-touch AI policies.
The framework proposed by OpenAI focuses on three goals: distribute AI-based prosperity more broadly, build safeguards to reduce systemic risk, and ensure broad access to AI capabilities to ensure that economic power and opportunity are not too concentrated.
OpenAI proposed shifting the tax burden from labor to capital. The company did not specify its corporate tax rate. During his first term, Trump cut the corporate tax rate from 35% to 21%. But OpenAI warns that AI-led growth could empty the tax base that funds Social Security, Medicaid, SNAP, and housing assistance as corporate profits expand and reliance on labor income decreases.
“As AI reshapes work and production, it could change the composition of economic activity, expanding corporate profits and capital gains while potentially reducing reliance on labor income and payroll taxes,” OpenAI said.
The company proposes higher taxes on corporate income, AI-based profits or capital gains. This is a policy category that led Marc Andreessen to support Trump after Biden proposed taxing unrealized capital gains in 2024. OpenAI also imposes a potential robot tax. This was proposed by Microsoft founder Bill Gates in 2017 and involves having robots pay the same amount of taxes to the system as their human replacements.
Tech Crunch Event
San Francisco, California
|
October 13-15, 2026
The document also includes a proposal to create a public equity fund to provide automatic public equity in AI companies and AI infrastructure, even if Americans do not invest in the market. All profits are distributed directly to citizens. This prospect may be appealing to Americans who have watched AI inflate markets without first-hand witnessing it.
Some of OpenAI’s proposals were labor-focused, including one to subsidize a four-day workweek without loss of wages. This is consistent with the tech industry’s promise that AI will provide humans with a better work-life balance. OpenAI also suggests that companies increase severance pay or endowments, cover a greater portion of health care costs, and provide subsidies for child or elderly care. In particular, OpenAI frames this as a corporate responsibility rather than a government responsibility, thereby excluding the people AI is most likely to replace. As jobs disappear due to automation, employer-sponsored health care and retirement benefits may also disappear.
That said, while OpenAI proposes separate portable benefits accounts that track workers across their careers, these would still likely rely on employer or platform contributions and fall short of universal, government-backed coverage that would fully protect those actually displaced by AI.
OpenAI acknowledges that the risks of AI go beyond job losses, including misuse by governments or malicious actors and the potential for systems to operate beyond human control. To mitigate these threats, we propose containment plans for dangerous AI, new oversight bodies, and targeted safeguards against high-risk uses such as cyberattacks and biological threats.
However, growth proposals are presented, including expanding power infrastructure to support AI’s power needs with safety nets and guardrails, and accelerating AI infrastructure construction by providing subsidies, tax credits, or equity. OpenAI says AI should be treated like a utility, and proposes that industry and governments work together to ensure that AI is affordable and widely available rather than controlled by a few companies.
OpenAI’s framework comes six months after rival Anthropic released a policy blueprint laying out a range of possible responses to AI-induced disruption.
“We are entering a new phase of economic and social organization that will fundamentally reshape work, knowledge, and production,” OpenAI wrote. This requires “a new industrial policy agenda that ensures that superintelligence benefits everyone,” the company says.
OpenAI was founded as a non-profit organization based on AI that benefits all humanity. It became a for-profit company last year, leading critics to question whether the company’s stated mission is compatible with its need to grow and fulfill its fiduciary duties to shareholders.
The company referred to previous eras of economic upheaval, such as the Industrial Age, and pointed out how new economic and financial movements such as the New Deal ensured that “growth translated into broader opportunity and greater security” by “building new public institutions, protections, and expectations about what a fair economy has to offer, including labor protections, safety standards, social safety nets, and expanded access to education.”
“The transition to superintelligence will require a much more ambitious form of industrial policy, one that reflects the ability of democratic societies to act collectively at scale to shape their economic future so that superintelligence can benefit everyone,” OpenAI said.
