Anthropic said Sunday that it will commit 5% of its shares to a newly created Public Benefit Fund, subject to stockholder approval, and estimates that roughly 20% of its stock will ultimately go to public causes once pledges from its co-founders and employees are counted. Elon Musk answered within a day: on Monday, he promised one SpaceX share for every child in the United States next year, a giveaway worth more than $10 billion at current prices.

The company behind the Claude chatbot cast the fund as a way to expand its philanthropic reach. “The Anthropic Public Benefit Fund will have increased flexibility and scale, enabling us to pursue our mission in new ways: for example, we might explore large donations to traditional nonprofits or direct distributions to specific populations,” Anthropic said in its announcement (Anthropic). The company has already committed $200 million to studying that approach.

The move lands amid public unease over the pace of artificial intelligence. An AP-NORC poll in late September found that 64% of U.S. adults say the technology is developing too fast, while 8% say it is moving too slowly.

Timing Around a Potential IPO

The fund takes shape as Anthropic prepares an initial public offering — a sale of shares to public investors — as soon as November, at a reported valuation of around $2 trillion. At that price, 5% of the company would be worth roughly $100 billion on paper, though that is the value of the shares rather than cash sitting in an account.

Structurally, the fund is set up as a wholly owned LLC rather than a nonprofit, an arrangement Anthropic says gives it room to switch between types of giving. The board can add another 2% of shares each time the company’s value rises 50% above a baseline that has not been disclosed, and none of the fund’s equity can be used for lobbying or political donations.

The corporate commitment sits on top of the co-founders’ pledge to donate 80% of their equity and a matching program for employee donations; Anthropic says roughly 20% of its shares in total will be dedicated to the public good. Open questions remain: the company has not said how the fund will be taxed or what it will disclose publicly, as a traditional charity must.

Musk: One Share Per Child

Musk, who has championed “Universal High Income” as distinct from Universal Basic Income, was quick to respond. His first reply to Anthropic, posted Sunday night, recommended “giving the shares directly to the public” instead of routing the equity through a fund (X). The pledge to America’s children followed early Monday, less than a day after Anthropic’s announcement.

“Next year, I will give one of my SpaceX shares to every kid in America. This will be the first of many.”
— Elon Musk (@elonmusk) October 19, 2026 (post on X)

With roughly 72 million children in the country, the pledge comes to about $10.5 billion at SpaceX’s price of roughly $147 a share. That is just over 1% of the 6.42 billion shares Musk held as of June 30, according to an SEC filing, and about the same slice of a fortune estimated at $908.5 billion — a figure that has crossed $1 trillion several times this year.

The giveaway is likely to run through Trump Accounts, tax-advantaged savings accounts for children that the Treasury Department recently opened to donations of individual company shares from companies and charities, with donated stock generally held for five years. That appears to be the option Musk intends to use instead of Anthropic’s fund-based path, which he more or less confirmed in a reply post on X:

“Thanks for helping enable this”
— Elon Musk (@elonmusk) October 19, 2026 (post on X)

A Safety Bill Days Earlier

The pledge comes from a company whose chief executive has been among the most prominent voices warning about AI risk. CEO Dario Amodei wrote in January that “we are considerably closer to real danger in 2026” (Decrypt). Anthropic’s positioning has also drawn fire from rivals: OpenAI CEO Sam Altman accused the company in April of using fear-based marketing to promote Claude Mythos (Decrypt). Anthropic is now giving away a slice of itself ahead of a stock market debut.

Two days before the fund announcement, Anthropic had quietly presented lawmakers with its own draft AI safety bill, as reported by Politico (The Daily Guardian). The draft would let the government block the release of a model posing a foreseeable, imminent risk of killing or seriously injuring more than 50 people or causing over $1 billion in damage — whether through a bioweapon, an AI slipping beyond human control, or a cyberattack. Any block would have to run through the Commerce Department and a federal court order, and developers would be required to publish safety frameworks and transparency reports for new models.

Anthropic says the draft reflects its Advanced AI Framework, which it has promoted to lawmakers since June — the same month Amodei called for binding rules on frontier models, meaning the most powerful AI systems (Decrypt).

Among the critics is David Sacks, an investor and former White House AI czar, who responded to the pledge with his own post on X: “You’ve put 20% of your equity in an LCC. Whether it serves the public good remains to be seen.” (post on X)

Anthropic says it will release the fund’s governance structure and valuation thresholds in the coming months. Its IPO, meanwhile, could come as early as November.