Anthropic commits 5% of shares to a Public Benefit Fund that grows with company value

The commitment

Anthropic has announced it will dedicate 5% of its shares to a new entity, the Anthropic Public Benefit Fund. The move is pending approval by stockholders, and the stake is designed to expand over time: the board may add a further 2% of shares for every 50% the company's value grows above a baseline that Anthropic says it will define later.

According to Anthropic, whose announcement reached the front page of Hacker News, the Fund is intended to extend the company's mission work beyond what it already does internally.

How the Fund is set up

Anthropic is a public benefit corporation, and much of its mission-oriented activity already happens inside the company: research on AI safety and interpretability, applied biomedical research, and discounted Claude access for nonprofits and scientific institutions. Because those activities carry both public benefit and commercial value, the company keeps them in-house.

The new Fund is meant to work differently. Anthropic describes it as having more room to operate at scale and to try new approaches. Structurally, it is a wholly-owned LLC rather than a US 501(c)(3) charity, a choice Anthropic says lets it move between different forms of giving. One restriction is explicit: equity held by the LLC cannot be spent on lobbying or political donations, in the US or abroad.

Anthropic also draws a contrast with common tech-industry practice. Companies routinely hand executives equity awards that swell as the valuation rises; Anthropic says it is pointing growth-linked equity of that kind at the public good instead.

What it might fund

One area Anthropic highlights is the labor market. The company argues AI could produce enormous economic value while destabilizing employment for many workers, and notes it has already committed $200 million to work on the problem. Given how fast AI is advancing, it says, more aggressive approaches may be needed, and the Fund is meant to have the latitude to try them.

Possible uses Anthropic names include sizable grants to established nonprofits and direct payments to particular groups of people. Further detail on governance, and on the valuation milestones that would grow the Fund's stake, is due in the coming months, along with a public benefit report on the company's progress against its mission.

Building on existing pledges

The commitment stacks on top of earlier ones. Anthropic's co-founders pledged at founding to donate 80% of their equity holdings, and the company operates a match program for employee equity donations in which roughly half of employees take part. Combining those with the new Fund, Anthropic puts the total at roughly 20% of its shares dedicated to public-good purposes, before any additional tranches triggered by valuation milestones.

Why it matters

Major AI labs face persistent skepticism about whether safety and public-benefit commitments are durable or merely rhetorical. Anthropic's move expresses part of that commitment as an ownership stake whose size follows defined valuation rules, and whose share of the company grows rather than shrinks as the company becomes more valuable.

Plenty of detail is still outstanding: the baseline valuation, who holds discretion over the milestone tranches, and how the Fund's spending will be defined and reported. Those answers, promised in the coming months, will shape whether this reads as a meaningful governance innovation or a symbolic gesture. But as an attempt by a leading lab to bind a slice of its future upside to the public interest, and to forgo executive-style escalating equity in the process, it sets a marker that other AI companies may face pressure to match.