Anthropic Public Benefit Fund: Musk Answers With a SpaceX Share for Every US Child

Anthropic will place 5% of its shares in the Anthropic Public Benefit Fund, an entity it owns in full, and the stake is designed to grow as the company’s valuation climbs. The artificial intelligence (AI) company announced the fund on 18 October, ahead of an expected initial public offering (IPO). Elon Musk replied the same day that the shares should go straight to the public, and the next day he pledged one of his own SpaceX shares to every child in America.

The exchange tests a promise AI executives repeat often, that the wealth the technology creates will be shared. It also raises the question of whether shares handed out at a private company’s discretion are the right way to share it. Both gifts rest on decisions their givers control, and in the United States the proposals that would make AI companies fund support for every worker are still bills in Congress.

What Anthropic is putting into the fund

Anthropic’s stockholders still have to approve the transfer of 5% of the company’s shares into the fund. The board can then add up to 2% more for every 50% rise in Anthropic’s value above a baseline the company has not yet disclosed. At the $965 billion valuation of Anthropic’s May funding round, the initial 5% is worth about $48 billion, and the fund arrives with the company’s IPO filing already made.

The fund will be a limited liability company owned entirely by Anthropic, a format the company chose over a 501(c)(3) charity so it can move between different kinds of giving. None of its Anthropic equity may pay for lobbying or political donations, in the United States or anywhere else. Anthropic counts the new stake alongside two older commitments: its co-founders pledged 80% of their equity when the company was founded, and roughly half of its employees donate equity through a matching programme. Together, those commitments put about 20% of Anthropic’s shares toward the public good, before any milestone top-ups.

An escalator borrowed from executive pay

Anthropic drew the comparison itself in its own announcement. Technology companies routinely grant executives equity awards that grow as the company passes value milestones, and Anthropic says it is redirecting that kind of stock toward public causes instead. The best-known award of that kind belongs to Musk: in November 2025, Tesla shareholders approved a package worth up to nearly $1 trillion, which would add as much as 12% of Tesla to his stake if the company reaches targets that include an $8.5 trillion market value by 2035.

We’re committing a portion of Anthropic’s shares to the Anthropic Public Benefit Fund, a new wholly owned entity dedicated entirely to the public benefit. It will start off with 5% of our shares. For every 50% increase in the company’s value over a set baseline, we can contribute up to an extra 2%, subject to our board’s discretion. In total, this means that roughly 20% of Anthropic’s shares are committed to the public good. Our co-founders have already pledged to donate 80% of their equity, and around half our employees are donating a portion of their equity as part of a donation matching program we’ve run since our founding. In the post below, we share more details about our new fund. https://t.co/XzDntVH0vH

— Anthropic (@AnthropicAI) October 18, 2026

The size of Anthropic’s escalator depends on the baseline it has kept private. A 50% rise from the May valuation would mean roughly $1.45 trillion, close to the $1.5 trillion figure Reuters reported some investors discussing ahead of the listing (The Bright Minded’s calculation, which assumes a baseline Anthropic has not confirmed). Anthropic says the fund could write large cheques to nonprofits or pay specific groups of people directly, and it names AI’s effect on jobs as one area. It has already earmarked $200 million for research and responses on that front, and says the pace of AI may call for bolder action.

Musk’s answer to the fund

Musk, whose businesses include a rival AI developer, responded directly under Anthropic’s post on X. He recommended giving the shares to the public instead, and the following day added that this is what he would do himself. His objection concerns where the shares sit: Anthropic keeps them in an entity it controls, and Musk wants them in the hands of individuals.

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

One SpaceX share for every American child

In his pledge on X, Musk said he would give one of his own SpaceX shares to every child in America next year and said more would follow. He gave no exact date, no delivery mechanism and no definition of who qualifies. Roughly 73 million Americans are under 18, according to Census figures cited by CNN, which puts the gift near $10.7 billion with SpaceX stock around $146 at Friday’s close. Musk disclosed about 4.77 billion SpaceX shares in his post-listing filing, so 73 million of them would be roughly 1.5% of his holding (The Bright Minded’s calculation).

The pledge also answers a request that has been waiting since July. President Donald Trump said in a CNBC interview in early July that he expected Musk to donate SpaceX stock to Trump Accounts, the tax-advantaged investment accounts for American children that launched on 4 July. Days later, SpaceX president Gwynne Shotwell and her husband pledged one share each to the accounts of more than two million children, prioritising 11 to 17 year olds in lower-income areas. Musk had not commented on it publicly at the time.

Where a gifted share would actually land

Moving single shares into tens of millions of hands is a financial technology problem before it is a philanthropic one. Trump Accounts are the channel Shotwell chose for her SpaceX gift, though Musk has not named them. The accounts were built to hold index funds, and the law behind them requires investments to track an index; the Treasury said around their launch that donated stock could go in, without setting out the process. Parents had signed up more than 6 million children by early June, a fraction of the 73 million Musk’s pledge would cover.

Anthropic’s version runs into a different constraint. Its shares do not yet trade on any exchange, which leaves a company-owned fund as the practical holder until they do. Any direct distribution the fund makes later, whether in cash or in stock, will need the same fintech plumbing Musk’s pledge depends on: accounts that can receive the asset, custody for minors, and identity checks that confirm who is eligible.

The promise of AI wealth, from policy to private discretion

AI’s leading figures have put the idea of shared AI wealth on the record well before this weekend. In March 2021, OpenAI chief executive Sam Altman published an essay proposing an American Equity Fund, financed by an annual 2.5% levy on large companies, paid in their own shares, and on privately held land, with dollars and shares distributed to every adult citizen. Its acknowledgements thank Daniela Amodei and Jack Clark, two of Anthropic’s co-founders. In April 2026, Musk pinned a post calling federal universal high income cheques the best answer to AI-driven unemployment, and OpenAI’s 2025 restructuring left its nonprofit foundation with a 26% stake in the company, worth about $130 billion at the time.

Altman’s plan and Musk’s universal high income both relied on public policy: a tax written into law in one case, federal cheques in the other. This weekend’s commitments run on private discretion. Anthropic’s first 5% still needs its stockholders’ approval and every later top-up sits with its board, while Musk’s pledge exists, for now, as a post on X.

What exists at national level

Washington has proposals that would make AI companies pay into schemes open to every worker. Senator Mark Kelly is backing the Make AI Work for Americans Act, which would require the companies gaining most from AI to fund a federal trust for workers, job seekers and their communities. In August, Representatives Greg Casar, Valerie Foushee and Sara Jacobs introduced the AI Tax and Work Protection Act, a levy on large AI companies whose rate climbs with unemployment and whose revenue would pay for a public jobs programme. Neither bill has passed.

Trump Accounts, the national programme that has drawn gifts of SpaceX stock, were not designed for AI’s effects on work. The federal government seeds $1,000 only for children born between 2025 and 2028, and the rest of each account depends on families, employers and donors such as Shotwell.

A gifted share adds to a family’s wealth while leaving that family without a say over how the company deploys AI, including in decisions that cut jobs. One share among several billion carries a negligible vote at SpaceX, where Musk holds roughly 85% of the voting power, and the shares in Anthropic’s fund stay inside Anthropic. With no national scheme yet built to share AI’s gains with everyone, the economic security of people exposed to AI depends on what each company decides to give, how much, and when.

Editor’s note

Every piece goes through careful verification, but mistakes can happen. Readers who spot an error or have additional information can write to rosalia@thebrightminded.com.