Anthropic Pledges 5% of Its Stock, and Growing, to a New Public Benefit Fund

Anthropic said Sunday it will commit 5% of its outstanding shares, subject to stockholder approval, to a newly created Anthropic Public Benefit Fund, a philanthropic vehicle the AI company says could eventually hold roughly a fifth of its equity as its valuation grows.

The fund is structured as a wholly owned limited liability company rather than a traditional nonprofit, which Anthropic says gives it flexibility to move between large grants to existing charities and direct cash payments to specific groups, such as workers displaced by automation. In its own announcement, the company said it is "instead directing that equity to the public benefit" rather than following the more common industry practice of awarding escalating equity grants to executives.

A Formula Built to Scale

The initial 5% commitment is only a floor. Anthropic's board can, at its discretion, add another 2% of shares to the fund for every 50% increase in the company's valuation above an undisclosed baseline, meaning the fund's stake is designed to grow in step with the company's own growth. Combined with two pre-existing commitments — Anthropic's co-founders pledged at the company's founding to give away 80% of their personal equity, and roughly half of Anthropic's employees take part in a separate equity-donation-match program — the company says total shares earmarked for "the public good" now stand at roughly 20%, a figure it expects to keep rising as the valuation-linked formula kicks in.

Anthropic was explicit about one limit on the new fund: none of its holdings will go toward lobbying or political giving, in the United States or abroad.

"None of the Anthropic equity held by the LLC will be used for lobbying or political giving, in the US or abroad," the company said in its announcement.

An Existing Governance Structure

Anthropic is organized as a Delaware public benefit corporation, a structure that requires its board to weigh a stated public-interest mission — "to responsibly develop and maintain advanced AI for the long-term benefit of humanity," per its charter — alongside shareholder returns. The company's existing Long-Term Benefit Trust, a panel of up to five financially disinterested trustees, is empowered to elect a growing share of Anthropic's board over time; former Federal Reserve chair Ben Bernanke joined the trust earlier this year.

The new fund is meant to sit alongside, rather than replace, mission-driven work the company already keeps in-house. Anthropic pointed to its existing $200 million Economic Futures Research Fund, which studies AI's effect on labor markets, and a $150 million Claude Corps fellowship program that places roughly 1,000 fellows with nonprofits, as examples of work the new fund could eventually scale up. The company drew an explicit line between that kind of giving and work it already does in-house that happens to carry commercial value too, such as AI safety and interpretability research, biomedical research, and discounted access to its Claude models for nonprofits and scientific institutions. The new fund, Anthropic said, is meant for efforts that do not fit that dual-purpose mold — giving with no direct business upside at all.

The numbers involved are difficult to pin down in dollar terms because they move with the company's own valuation, which is itself unsettled. If Anthropic's valuation were to approach the roughly $2 trillion some reports have floated around its planned share sale, a fund holding a fifth of the company's equity would be worth several hundred billion dollars on paper — a figure that would rise and fall with Anthropic's share price rather than representing a fixed cash pledge today.

An IPO in the Background

The announcement comes as Anthropic prepares for what Reuters and other news organizations have reported could be one of the largest initial public offerings in history, with some reporting pointing to a share sale that could value the company near $2 trillion and a listing as early as November. Anthropic has not confirmed those figures itself, and the timing remains unsettled across multiple reports.

Coverage from TechCrunch framed the move against a backdrop of rising public skepticism toward the AI industry. Anthropic's own statement pointed to the scale of the challenge it says is ahead: "Given the pace of AI progress, bolder approaches may be needed. The Fund will have the versatility to pursue them." The news was also reported independently by Reuters and by the Associated Press, both of which confirmed the deal's terms directly with the company.

What's Still Unknown

Anthropic says it will publish details on the fund's governance, its decision-making structure and the valuation thresholds that trigger additional contributions "in the coming months," alongside a forthcoming public-benefit report. The fund itself still requires stockholder approval, which the company has not yet disclosed obtaining.

No Anthropic executive was quoted by name in the announcement or in the wire coverage reviewed for this article; every quote attributed to the company came from its corporate statement rather than an individual. Nor had any outside reaction — from labor economists, nonprofit groups or AI critics — surfaced in the hours after the announcement, a gap likely to fill in as the news spreads through an industry already arguing over how AI companies should be judged on more than their technology.

The move also puts a number on an argument Anthropic has made in other contexts: that the usual Silicon Valley playbook, in which a fast-growing company's biggest equity grants flow upward to founders and executives, is not the only way to allocate the wealth created by an AI boom. Whether rivals follow with similar commitments of their own, or whether the fund proves durable once outside scrutiny of its governance and payouts begins in earnest, is likely to be the real test of Sunday's announcement.