Bernie Sanders

“No longer would the future of A.I. and the transformation of human life that it will bring be dictated by a handful of Big Tech oligarchs,” Sanders recently wrote in The New York Times. “The federal government would have the power, through its voting shares and an equal representation on each company’s board, to block decisions that hurt our citizens and to push for policies that help them.”

The government would use these shares to build what’s known as a sovereign wealth fund, which could be used to improve the lives of Americans. As AI companies become more and more profitable, the public could reap the benefits. Sanders suggests that this revenue source could be used to make “direct payments to the American people” and help fund a stronger social safety net, from housing to education to health care.

This may sound like a pie-in-the-sky proposal. But it has bipartisan appeal. President Trump has suggested that he’s open to something similar. And according to the news site NOTUS, Sam Altman, the CEO of OpenAI, spoke with Trump about the possibility of AI companies voluntarily ceding some of their shares to the government, which would use the return to fund public benefits, such as direct payments to households.

But is Sanders’s idea as good as it may sound?

There are aspects of the senator’s proposal that have the potential to succeed. Owning half a given company’s shares would give the government much more say in that company’s direction — helping to shape its goals so that both the company and its products are more closely aligned with the public interest.

But there’s a contradiction at the heart of Sanders’s proposal.

The senator says a government stake in AI companies would allow it to restrain the firms. But as Nathan E. Sanders, a data scientist affiliated with the Berkman Klein Center at Harvard University, points out, a government that relies on AI wealth to pay for a variety of programs would have good reason to deregulate the industry it’s invested in so that it can maximize profits.

Sanders (no relation to the senator) cites several instances where this sort of conflict of interest has come up. The Norwegian government’s stakes in oil companies, for example, are not necessarily “steering the behavior of those corporations to align with what you might expect the values of the Norwegian public to be with respect to fossil fuel usage and climate change and carbon emissions,” he says. Instead, there’s evidence that those stakes seem to be inhibiting “the ability of the Norwegian government to implement those types of [environmental] policies because they also need to look out for the economic interests that they have directly in those companies through the sovereign wealth fund.”

In the same way, if the US government starts to depend on a sovereign wealth fund that’s tied to the success of AI companies, then the government might not be inclined to, say, hold back potentially harmful technologies that could generate significant revenue.

“The fact that there does seem to be support … from some of the AI companies themselves for this solution should give us some pause,” Sanders says. “Why is it that a company would be open to voluntarily, perhaps, ceding some significant fraction of its stock to the government, to the public?”

This is not to say that Senator Sanders’s plan should be dismissed outright. Having the government own shares in these companies can help democratize the governance of the AI industry. But intertwining that goal with the creation of a sovereign wealth fund that’s entirely predicated on the profitability of AI could have unintended consequences like an all-too-friendly regulatory environment. Plus, the government has other levers at its disposal, including higher taxes and the power to regulate the industry and influence corporate behavior.

Ultimately, lawmakers should keep in mind that any kind of regulation or governance of AI ought to be as insulated as possible from political and economic pressure — because when it comes to this particular industry, the welfare of humanity is on the line.