OpenAI co-founder and CEO Sam Altman has outlined a broad policy vision for the age of superintelligence, arguing that society will need new economic and social arrangements if artificial intelligence begins to outperform humans at the highest level. In the proposal, Altman frames the issue not simply as a technological milestone, but as a structural turning point that could reshape work, taxation, public benefits, and access to opportunity.
The backdrop is OpenAI’s description of superintelligence as AI systems capable of outperforming the smartest humans even when those humans are assisted by AI. If such systems continue to advance, Altman suggests that governments, companies, and citizens will need to prepare for far more than incremental labor-market disruption. The central challenge, in his view, is how to preserve social stability while allowing the productivity gains from AI to benefit society more broadly.
A Public Wealth Fund to Share AI Upside
One of the most notable elements in Altman’s proposal is the creation of a public wealth fund developed in partnership with AI companies. Under this idea, every citizen would hold a direct stake in a vehicle that invests in technology and AI businesses. The goal is to ensure that the economic gains generated by advanced AI do not accrue only to company founders, shareholders, and a narrow set of firms, but instead flow outward to the broader population.
This concept reflects a larger concern now surfacing in AI policy debates: if artificial intelligence dramatically lifts productivity, who captures the value? Altman’s answer is that a portion of the upside should be socialized through a public mechanism that gives ordinary people an economic claim on the expansion of the AI sector. Rather than relying only on wages as the primary route to prosperity, the proposal points toward ownership and capital participation as part of the future policy mix.
From Labor Taxes to Corporate Profits
Altman also argues that the tax system may have to adapt if AI leads companies to reduce payrolls while increasing output. In many existing systems, social benefits are funded in part through labor-related taxation. If fewer workers are needed, or if wage growth weakens relative to profits, then programs tied to payroll-based contributions could come under pressure.
For that reason, the proposal supports shifting the tax focus from labor toward corporate profits. Altman presents this as a practical response to the possibility that AI-heavy business models may generate large gains with a smaller human workforce. If that happens, traditional tax structures may no longer align with the real sources of value creation in the economy.
He specifically links this concern to the sustainability of major public programs such as Social Security and Medicare. In other words, the proposal is not just about redistributing wealth in a general sense; it is also about protecting the fiscal foundations of the social safety net if employment patterns change substantially. The article notes that similar ideas have also been supported by Anthropic CEO Dario Amodei and former U.S. presidential candidate Andrew Yang, indicating that the issue is becoming part of a wider policy conversation around AI.
A 32-Hour Workweek Without Pay Cuts
Another major recommendation is a shorter workweek. OpenAI’s proposal calls for reducing work time to 32 hours per week without a loss in pay, on the premise that rising AI-driven productivity should translate into better conditions for workers rather than only higher corporate margins. If machines and software systems can do more of the economically valuable work, then people, according to this logic, should be able to work less while maintaining living standards.
The proposal also calls for increased contributions to retirement and healthcare funds, suggesting that the AI transition should be managed with stronger long-term protections rather than weaker ones. This reflects a broader principle running through the document: technological acceleration should not automatically erode worker security. Instead, policy should aim to convert productivity gains into reduced strain, more resilience, and a more durable social contract.
While the idea of a shorter workweek has circulated for years in labor and economic debates, Altman’s intervention is notable because it ties the concept directly to the expected consequences of advanced AI. In this framing, the question is no longer whether society can imagine working less in theory, but whether institutions can evolve quickly enough to keep pace with the productivity shock that AI may produce.
AI as a Basic Right
Beyond income and taxation, Altman’s proposal argues that access to AI should be treated similarly to access to electricity or the internet. The document says individuals should have basic access to a certain level of AI capability so they are not left behind as the technology becomes embedded in education, communication, work, and economic participation.
This idea positions AI not merely as a commercial product, but as a foundational layer of modern life. If advanced tools become essential for learning, job seeking, decision-making, and entrepreneurship, then exclusion from AI access could become a new form of structural disadvantage. Altman’s framework therefore treats baseline AI availability as part of the infrastructure needed for an inclusive digital economy.
In parallel, the proposal calls for basic safety nets and guardrails against AI-driven job displacement. The aim is to ensure that workers who are most affected by technological shifts can receive support when the disruption is most severe. This is consistent with the broader message of the document: AI progress may be powerful, but it should not be managed as though social consequences are secondary.
Warnings on Cyber and Biological Risks
Altman’s concerns are not limited to economics. He also raises the prospect that superintelligent systems could be used in cyberattacks and biological threats, including the engineering of new pathogens. Those warnings place the proposal in a dual context: AI is both a source of extraordinary productivity and a potential amplifier of systemic risk.
According to the report, Altman told Axios that he wants serious debate on these issues to begin in earnest. That remark underscores a key point in the document’s tone. The proposal is not framed as a distant or abstract thought experiment. Instead, it is presented as an early attempt to prepare for capabilities that may have profound consequences for labor markets, public finance, public safety, and democratic governance.
A Broader Political and Economic Debate Ahead
Altman’s “new AI deal” does not offer a complete legislative blueprint, but it does lay out a coherent policy direction: broaden ownership of AI upside, redesign taxation around where value is actually created, reduce work hours as productivity rises, protect retirement and healthcare systems, guarantee baseline access to AI, and build safeguards for both labor displacement and catastrophic misuse.
That combination reflects an increasingly visible shift in the AI debate. The conversation is moving beyond model performance and product launches toward questions of distribution, governance, and institutional adaptation. If superintelligence becomes a serious policy horizon rather than a speculative concept, proposals like this may shape how governments and industries think about the next social contract.
For now, the significance of Altman’s intervention lies less in immediate implementation and more in the fact that one of the most influential executives in AI is openly arguing that economic rules, labor norms, and public protections may need to change alongside the technology itself. As AI capabilities accelerate, that argument is likely to gain even more attention.