OpenAI's new social contract: how the company wants to reshape society for the AI age

OpenAI published a sweeping industrial policy document this week calling on the U.S. government to prepare for a world where artificial intelligence outperforms the smartest humans and where the economic fallout could shatter existing safety nets if nothing is done.

The document, titled Industrial Policy for the Intelligence Age: Ideas to Keep People First, frame the moment as the beginning of a transition to “superintelligence”, AI capable of outperforming even AI-assisted humans, and argue that the current policy toolkit simply isn't built for what's coming.

The transition to superintelligence is not a distant possibility — it's already underway, and the choices we make in the near term will shape how its benefits and risks are distributed for decades to come.

The boldest ideas on the table

The document proposes a sweeping menu of policy experiments, with some landing far to the left of what most tech executives have been willing to touch publicly.

Why is OpenAI saying this?

The cynical read: a company valued near $300 billion asking governments to tax and regulate its industry has a self-interest in shaping what that regulation looks like. Getting ahead of the conversation means influencing the outcome.

The less cynical read: the CEO of a frontier AI lab is genuinely warning that his own technology could break the economy, and that doing nothing is the riskier bet. The document is careful to note these are "intentionally early and exploratory" ideas, not fixed recommendations, and it invites public feedback at a dedicated email address.

Either way, it's the most detailed public blueprint any major AI company has published on redistribution, taxation, and worker protection in the context of advanced AI. That alone makes it worth paying attention to.

The two-track framework

The document splits its ideas into two buckets: building an open economy (sharing the upside) and building a resilient society (managing the downside). On the economic side, it focuses on portable benefits, adaptive safety nets that auto-expand during disruption, and modernising the tax base as payroll taxes shrink. On the resilience side, it calls for AI auditing regimes, incident reporting systems, mission-aligned corporate governance, and international coordination bodies modeled on aviation or food safety regulators.

People are asking

Is this actually going to happen?

Almost certainly not in full, and probably not fast. The document explicitly frames these as conversation starters, not legislative proposals. But ideas like a public wealth fund and robot tax have been circulating in academic and policy circles for years, having OpenAI put its name on them shifts the Overton window.

What's the Alaska Fund comparison about?

Alaska's Permanent Fund pays every resident an annual dividend from oil revenues. OpenAI is proposing a similar structure , seeded by AI companies and invested in diversified assets , so that people who never own AI stocks still share in AI-driven growth.

What does "superintelligence" actually mean here?

OpenAI defines it as AI systems that can outperform the smartest humans even when those humans are themselves assisted by AI. They're not claiming it's arrived, they're saying the transition has begun and policy needs to move now.

Is this just lobbying dressed up as policy?

That's a fair question. The document does call for targeted regulation only on the most advanced models, i.e., companies like OpenAI, which could entrench incumbents as much as constrain them. The call for public input and research fellowships ($100K grants, $1M in API credits) suggests genuine outreach, but OpenAI is still steering the conversation.