AI founders want to give shares to the public? Give them to Social Security.

Anthropic, readying an initial public offering that investors reportedly value at $1.5 trillion to $2 trillion said it will commit 5 percent of its shares, subject to shareholder approval, to a new Public Benefit Fund, with up to 2 percent more for every 50 percent gain in the company’s value. With earlier pledges, roughly a fifth of the company is earmarked for the public good. 

Elon Musk’s response: Give the shares directly to the public. On Oct. 19, he pledged to do it himself: one SpaceX share, currently worth about $146, to every American child next year.

He is closer to right than he knows. The country already has an institution that insures nearly every worker against lost earnings and pays those workers directly: Social Security. But Social Security is not allowed to own equity shares; by law it may hold only federal debt. 

Congress should build an independent fund that accepts contributed equity from any company, holds it in a diversified portfolio and dedicates the returns to the trust funds. Canada’s pension plan, Norway’s oil fund, Alaska’s Permanent Fund and state pension systems own equity on the public’s behalf. To allay any fears, the fund’s governance should be boring by design, with no government board seats, no political direction and a published schedule for selling each stock down into a broad index. 

Anthropic and Musk are not alone in wanting the public to share in AI’s gains. OpenAI’s April policy paper proposed a public wealth fund investing in AI companies and paying returns to citizens. In July, the Financial Times reported Sam Altman had offered 5 percent of the company’s equity to seed a sovereign........

© The Hill