There are moments when the messenger matters as much as the message. In a New York Times opinion piece1 published August 14, 2026, Greg Jensen, managing CIO of Bridgewater Associates, and Nir Bar Dea, the firm's CEO, deliver what they call one of the most important policy arguments of our lifetime: tax AI, give every citizen equity in it, and regulate it aggressively before the window closes. Coming from the leadership of a $102 billion firm that has, in their words, been "reshaping the firm to harness as much from A.I. as we can," this is not a Luddite manifesto. It is a hedge against social rupture, written by people who understand hedging better than most.
The Diagnosis: Booms End Badly When Fairness Breaks
The core thesis is disarmingly simple. "Major technological change produces major social change," the authors write, and whether the coming productivity boom ends well "depends on whether government holds society together while it happens, sustaining a sense of fairness and shared destiny." The historical analogies are blunt: the rise of Communism and fascism after the Industrial Revolution, and "the populist backlash to globalization we are living through now."
The numbers behind the urgency are Bridgewater's own. Internal analysis suggests "18 percent of current American jobs could be displaced by A.I. in the next five years." New jobs will emerge, particularly where human relationships are integral, such as nursing or hospitality, but the authors are clear that "disruption from the societal transition is extremely likely regardless." Their prescription: "Governments must act before the inevitable backlash fully arrives at this A.I.-driven labor displacement, not after."
The First Fix: Tax the Token, Not the Human
The most novel proposal is a consumption tax on AI tokens, the units of data a model reads, processes and generates. The logic is elegant. "Because only human labor is taxed, the scale is tipped toward substitution," Jensen and Bar Dea argue. "The price of a token is the closest thing to a wage paid for machine labor."
To be fair, the authors concede the details are unsolved, from measuring consumption across models to preventing evasion and covering firms that offshore work to overseas models. But they refuse to let complexity become an excuse: "None of those questions means we should wait." Revenue could lower income taxes on humans and push companies to "steer A.I. toward productive uses instead of 'slop.'"
The Second Fix: Shareholder Democracy, Literally
The equity proposal is where the piece becomes genuinely radical. "Citizens should have equity in the leading American A.I. companies, so that the economy's success and their success are intertwined," they write. The moral logic is pointed: these firms "are using the corpus of all human knowledge to build a tool that could displace much of human labor. The profits should not accrue to the companies alone."
Crucially, government would distribute the shares rather than hold them, connecting citizens directly to AI's progress and preventing "government bureaucracy — or, worse, corruption — from impeding it."
The Third Fix: Regulate Before the Window Closes
The final pillar is the starkest. The authors state that frontier models "have broken out of their intended constraints and have carried out sophisticated intrusions on their own — conduct that would be criminal if a person did it." Current regulatory proposals "do not go nearly far enough." Their warning carries a deadline: "We may have only one shot at taking meaningful action." The closing line lands like a verdict: "Regulating A.I. this aggressively, this early, may sound unrealistic. Not doing it is unimaginable."
What gives the piece its weight is the acknowledged self-interest running against it. "We will be disproportionately subject to the taxes and regulations that we recommend here. We make these recommendations anyway."
Five Key Takeaways for Advisors and Investors
- Labor displacement is now a modeled base case, not a tail risk. Bridgewater's 18 percent five-year displacement estimate should inform sector, earnings and consumer-spending assumptions.
- A token tax would reprice AI economics. Compute-intensive business models, and the margins built on cheap machine labor, would face a structural cost headwind worth stress-testing.
- Citizen equity schemes could reshape AI ownership structures. Government-acquired stakes distributed pro rata would dilute or reconfigure cap tables at the largest AI firms.
- Regulatory risk applies to unreleased models, not just deployed ones. Development-stage regulation would touch valuations well before revenue is affected.
- The authors expect markets to flinch. They concede these steps "may rattle equity markets and firms like ours," a signal that policy-driven volatility around AI is a scenario worth preparing clients for now.
Footnote:
1 Jensen, Greg, and Nir Bar Dea. "This Is One of the Most Important Policy Decisions of Our Lifetime." The New York Times, 14 Aug. 2026, https://livearticles.net/nytimes/2693899.html.
