We tend to think that OpenAI and AI companies in general push for minimal regulation so that nothing stops their expansion. But in a new document, “Industrial policy for the age of intelligence” OpenAI argues the opposite: that government oversight is not only necessary; we need it now.
He suggests that we need a new kind of industrial policy to deal with the upheaval that AI will create: “The transition to superintelligence will require an even more ambitious form of industrial policy, reflecting the ability of democratic societies to act collectively, at scale, to shape their economic future so that superintelligence benefits everyone.”
The risk to employment is presented as the most immediate threat. “While we firmly believe that the benefits of AI will far outweigh its challenges, we are clear-eyed about the risks: that jobs and entire industries will be affected,” the document says. It also points to broader concerns, including threats to democracy, the concentration of wealth in the hands of a few, and the misuse of technology by “bad actors.”
Article continues below.
OpenAI suggests expanding what it calls the “care and connection economy” – roles in child care, elder care, education, healthcare and community services – as new avenues for workers. It also proposes turning AI-driven efficiency gains into tangible benefits for employees.
That’s where one of his most striking ideas comes into play.
OpenAI suggests that governments and employers should “incentivize employers and unions to pilot time-bound 32-hour/four-day workweeks, with no loss of pay, that keep production and service levels constant.” If successful, those shorter weeks could become permanent or translate into more paid time off.
It’s an attractive vision: AI does more work and humans get more time back. But it’s based on a big assumption: that those productivity increases actually translate into higher wages, not just higher profits.
Tax the rich
The OpenAI paper also hints that a more fundamental change is needed in the way economies are taxed. If AI reduces the need for human labor, he argues, governments may need to rely less on taxing workers and more on taxing the capital and businesses that benefit most from automation. In other words, if AI does more of the work, companies that benefit from it may have to shoulder more of the burden.
One of the most radical proposals contained in the document is the idea of a public wealth fund. It is effectively a way to redistribute the benefits of AI throughout society. Instead of those profits staying in the hands of a handful of companies, they could be pooled and returned to citizens, echoing models like sovereign wealth funds.
However, the most surprising thing about this document is how OpenAI frames AI itself. It’s no longer just about better chatbots or smarter tools. The document positions AI as closer to infrastructure: a fundamental layer that will underpin entire industries, economies, and public services.
The future is bright
OpenAI argues that AI could reshape the economy for the better, while outlining how easily it could concentrate wealth and power if left unchecked. History has told us that humans are generally bad at long-term planning, and it will take a very different mindset than what our political leaders currently embody to implement these ideas.
It’s also an unusual position for OpenAI: the company that builds AI technology is now helping to outline how AI should be taxed, regulated and redistributed. Whether that’s forward-thinking or selfish will depend on who you ask, although the idea of a four-day work week would get my vote.
The best business laptops for every budget
