Anthropic, the company behind the Claude models, announced the Anthropic Public Benefit Fund on Sunday, October 18, 2026. It is committing 5% of its shares to the fund now, subject to stockholder approval. The board can add another 2% for every 50% increase in the company’s value above a baseline that Anthropic has not made public.
Anthropic’s post puts no dollar value on those shares. At the $965 billion valuation set in its Series H round, announced May 28, 2026, that 5% stake is worth about $48 billion. That is a paper value for shares in a private company. No payment has been made, and the commitment still needs the stockholder vote.
According to Anthropic, the fund could make large donations to nonprofits or “direct distributions to specific populations.” The only example it gives is AI’s effect on jobs.
A stake that can grow with the valuation
The mechanism has two tiers. The first is fixed: 5% of the company, committed on October 18 pending the stockholder vote. The second depends on growth: another 2%, “at the board’s discretion,” for every 50% increase in value above a baseline. “Because of the way the Fund is structured, the more Anthropic grows, the larger the share of the company that can be dedicated expressly to the public good,” Anthropic writes.
The company contrasts this with industry practice: “It is common practice for tech companies to offer escalating compensation awards to their executives. We are instead directing that equity to the public benefit.”
The fund is an LLC wholly owned by Anthropic. Anthropic says it chose that form over a 501(c)(3) charity “to give adjustability between different forms of giving.” It pledges that none of the Anthropic equity held by the LLC will be used for lobbying or political giving, in the US or abroad.
| Status | What it covers |
|---|---|
| Committed, subject to stockholder approval | 5% of Anthropic’s shares |
| At the board’s discretion | 2% more per 50% increase in value above an undisclosed baseline |
| Promised “in the coming months” | More on the fund’s structure, governance and valuation milestones |
| Announced, no date | A public benefit report on progress against the mission |
Donations, direct distributions and jobs
Founded in 2021 by former OpenAI employees, Anthropic is a public benefit corporation incorporated in Delaware: its directors must balance returns to shareholders with the mission written into its charter. According to the post, most of its mission work happens inside the company. That covers AI safety and interpretability research, applied biomedical research, and discounts on Claude for nonprofits and scientific institutions.
The fund is meant to go further, with “increased flexibility and scale.” Anthropic writes that it “might explore large donations to traditional nonprofits or direct distributions to specific populations.” The post does not say which populations, what form the distributions would take, or when.
The example it picks is the labor market. “AI may generate enormous economic value, but it could also cause instability for many workers,” the company writes. It notes that it has “already committed $200 million to studying and addressing this issue,” and adds that “bolder approaches may be needed.” US taxi and ride-hail drivers could become one concrete case: Waymo plans to grow from 4,000 to 20,000 robotaxis by the end of 2027, unnamed sources told Semafor (see our article on Waymo’s crash data and fleet plans).
That $200 million is the Economic Futures Research Fund, whose research agenda Anthropic published on July 22, 2026. The fund plans to back outside research, mostly projects of $5 million to $30 million, across five areas. Two of them are modernizing income support for workers displaced by AI and building worker stakes in AI-driven growth before disruption arrives.
Anthropic describes that research fund as global, while acknowledging its funding directions are “somewhat US-centric.” The company also publishes its own hiring data, such as the share of junior employees among its technical hires, which fell from 27% to 9%.
Earlier pledges and the roughly 20% total
The fund adds to older commitments. When Anthropic was founded, its co-founders pledged to donate 80% of their equity. The company also matches equity that employees pledge to donate, a program about half of them take part in, according to the post.
Combining those pledges with the new fund, Anthropic says “roughly 20%” of its shares will be dedicated to the public good, plus any milestone additions. Of that 20%, only the fund’s 5% is new, and the post does not break down how much each part contributes.
An announcement ahead of the listing
The October 18 post does not mention a stock listing, but the press ties the announcement to the IPO Anthropic is preparing. The Guardian reports that the company is considering a $1.5 trillion valuation, with an IPO planned for November. Al Jazeera writes that it is expected to list in the US before the end of 2026, with analysts forecasting a valuation above $2 trillion. By our calculation, assuming the share count stays the same, a 5% stake would be worth $75 billion at the first figure and over $100 billion at the second.
According to a Reuters dispatch carried in French by Boursorama, the main body of Anthropic’s prospectus runs 261 pages, about 80 of them on risk factors. The Guardian places the announcement amid growing public nervousness about AI, from jobs to the environment. The paper also notes that Anthropic warned the previous week that some of its models could soon copy themselves onto connected devices.
As reported by The Guardian and Reuters, Elon Musk, the Tesla chief executive and SpaceX founder, replied on X: “I recommend giving the shares directly to the public. This is what I will be doing.” He later added: “Next year, I will give one of my SpaceX shares to every kid in America. This will be the first of many.” The Guardian also writes that OpenAI reportedly held early talks about giving a 5% stake to the US government.
The conditions and the open questions
The $48 billion figure does not appear in Anthropic’s post: it comes from the May 2026 private valuation. If the company lists, the value of the shares will depend on the market price. No donation has been made. The 5% awaits the stockholder vote, and each 2% tranche depends on an undisclosed baseline and on a board decision.
The LLC belongs to Anthropic. The post does not say who will decide how the fund is used, or under what rules. It also does not say whether beneficiaries outside the US are being considered, and no program has been announced in Europe. “Direct distributions” remain an idea: Anthropic writes that it “might” explore them, and it announces no payments.
What it means
The announcement ties part of the equity of one of the leading AI developers to public-benefit uses, and allows that part to grow with the company’s value at the board’s discretion. The example it chooses, workers affected by AI, matches the subject of its $200 million research fund. For businesses that use Claude, the post announces no change to plans or prices.
The gap between the announcement and what is settled remains wide. The amount cited in the press is a paper estimate, the fund is wholly owned by Anthropic, and no beneficiary is named. Elon Musk argues for a different route: handing shares directly to the public. If confirmed, the reported OpenAI talks with the US government would show other companies weighing ways to share equity.
Several steps will show the fund’s real reach: the stockholder vote, the publication of its governance and baseline, the promised public benefit report, then the first donations or distributions. How the listing documents describe the fund will be another signal.
Sources
- Investing in the public benefit, Anthropic, October 18, 2026
- Anthropic raises $65B in Series H funding at $965B post-money valuation, Anthropic, May 28, 2026
- A research agenda for the Economic Futures Research Fund, Anthropic, July 22, 2026
- AI firm Anthropic to commit 5% of shares to fund for ‘public good’, The Guardian, October 19, 2026
- AI giant Anthropic commits 5% of shares to ‘public benefit’ fund, Al Jazeera, October 19, 2026
- Anthropic va céder 5 % de son capital à un nouveau « fonds d’intérêt général », Reuters via Boursorama, October 19, 2026
Header image: MACCUS chart, data from Anthropic (May 28, 2026), The Guardian and Al Jazeera (October 19, 2026)
