Anthropic’s latest pledge begins with a striking number: subject to shareholder approval, the company plans to commit 5% of its shares to a new Anthropic Public Benefit Fund. At its most recent reported valuation, that stake would be worth roughly $48 billion.¹

The company announced the proposal as an expansion of a mission it says already includes AI-safety research, biomedical work and discounted access to Claude for nonprofits and scientific institutions. The fund would sit alongside those efforts as a wholly owned LLC, rather than a conventional charity, a structure Anthropic says permits more “adjustability between different forms of giving.”²

Its promised scale could rise with the company’s value. Anthropic says the board may add another 2% of shares for every 50% increase above a future baseline. It frames that approach as a deliberate alternative to escalating executive equity awards: “We are instead directing that equity to the public benefit.”²

The proposed uses are broad. Anthropic has floated large grants to established nonprofits or direct distributions to particular populations, while ruling out lobbying and political giving. Its clearest example is the labor market: AI may create vast value, it says, but could also bring instability for workers. The company has already committed $200 million to research and address that risk, arguing that faster AI progress may demand “bolder approaches.”²

That flexibility is also where the public-interest case becomes harder to assess. Reporting on the announcement notes that Anthropic has supplied only limited detail on how the money will be used, while another account says the company has yet to identify who will decide which projects qualify for support.³

Anthropic says governance and valuation milestones will be disclosed in coming months. Until then, the pledge offers an unusually large potential pool for public benefit—but leaves control of that pool firmly inside the company.²