The giant hedge fund Bridgewater Associates has emerged as an unlikely voice for the little guy in the AI debate, calling for protecting everyday Americans from the dangers of the technology — and spreading the wealth that tech giants are reaping from it.
The firm founded by billionaire Ray Dalio — famous for the insular, secretive culture at its posh Connecticut headquarters — has held talks with Washington policymakers from both parties about proposals to avert an AI-driven job apocalypse, The Post has learned.
Last month, Bridgewater’s chief investment officer Greg Jensen, CEO Nir Bar Dea and others published a lengthy essay that called for “immediate policy action” in order to realize the “full potential of artificial intelligence by mitigating the risks of widespread societal disruption and catastrophic safety accidents.”
The concepts have caught the attention of lawmakers on Capitol Hill and have spurred informal talks with Bridgewater about the need to create “citizen equity” in the AI boom, a source told The Post.
Among its proposals is a so-called “token tax” on AI usage. The revenue from a 35% tax could reach $600 billion by 2030 and could be used to acquire shares of AI giants and distribute them to members of the public, according to the paper.
Bridgewater’s token tax contrasts with calls for “universal basic income” from tech luminaries including OpenAI CEO Sam Altman and SpaceX CEO Elon Musk – a form of social welfare where every citizen receives a set, recurring payment from the government. In a Bridgewater Q&A, Jensen said UBI would hand too much control to the government.
“You’re still dependent on a bureaucrat deciding how much and when you get your check,” Jensen said. “If we actually distribute the equity to all citizens it once again takes the power away from the politicians to use that in ways that you might not want and it gets power directly to the citizens.”
Jensen and Bar Dea wrote in the report that their proposals are coming now because time is running out to ensure everyday Americans get to share in the upside of the AI boom.
“Absent intervention today, mitigation of that danger will become nearly impossible as diffusion of the technology accelerates and models themselves become capable of improving and acting autonomously.”
Ensuring smooth adoption of AI could ultimately benefit Bridgewater’s own investments in the sector by limiting long-term public backlash and stemming more onerous restrictions on the tech, AI observers say.
Jeremy Bearer-Friend, an associate law professor at George Washington University, lauded Bridgewater’s proposals, telling The Post that one of the overlooked benefits of a public equity stake in AI is its benefit to public safety.
“Public equity could give public voice within traditional corporate governance structures on top of the needed regulation of the sector,” Bearer-Friend said. “An AI equity tax could give the public a role in boardroom decisions about public safety.”
Bridgewater says AI-generated work is increasingly substituting for human labor, even though human workers and employers remain subject to payroll taxes. The firm said proceeds from a token tax could be used to cut taxes on human workers and help people who are put out of work by AI.
“I think it just is common sense that we don’t want to incentivize machine labor over human labor,” Jensen said in an internal Bridgewater Q&A about the proposal, adding that a division within the Internal Revenue Service should be created to enforce it.
Jensen took his regulatory calls a step further in subsequent comments this month to the news site The Information, arguing that AI giants that control more than, say, 5% of US or global AI computing power, ought to receive heightened oversight, similar to how banking heavyweights with sway over the economy receive increased scrutiny.
“In two years, OpenAI and Anthropic are going to control 35% to 50% of the world’s compute. That’s a crazy outcome for a society to allow on something as powerful as compute,” Jensen said. “Would we let one entity control that much of some other form of energy or commodity?”
“I’m no big fan of government regulation, but if the problem’s big enough and society bears the risk, you need government regulation,” Jensen told the Information.
Bridgewater said in it’s post that it would be “disproportionately subject to the taxes and regulations we recommend” and was promoting the policies “in order for the benefits of AI to be realized in the long run.”
Bridgewater also called for a more active government role in policing AI safety, with regulators “regularly conducting sworn (subject to penalty of perjury) interviews of AI lab staff members about safety risks in emerging models and steps the labs are taking to address those risks,” the firm wrote.
Meanwhile, Bearer-Friend cautioned against “giving AI companies even more public funding by requiring government to purchase AI shares.”
Cyril Gorlla, founder and CEO of AI startup CTGT – which advises clients including big banks on AI – said Bridgewater’s idea to spread the AI wealth and bolster safety could help instill confidence in AI, further fueling the industry’s growth.
Gorlla added that implementation could be tricky, particularly in regards to a token tax.
“Jensen mentions that they basically have to build a way to collect those taxes which I think is fundamentally correct,” Gorlla told The Post. “But there is no universal metering standard with AI like you have with utilities where you can walk out and you have a meter box.”