Anthropic proposed committing 5% of its shares to a new Public Benefit Fund on Sunday, subject to stockholder approval. The Public Benefit Fund is being established as a wholly owned limited liability company. Anthropic says the fund could support donations to nonprofits or direct distributions to specific populations, including efforts to address AI's effects on workers.

Anthropic's separate $200 million Economic Futures Research Fund supports external studies of responses to AI's economic effects. Its research agenda was published on July 22, before the new fund proposal.

What Changed

AI-generated summary, reviewed by an editor. More on our AI guidelines.

A stake that could grow

The board could add another 2% of Anthropic's shares for every 50% increase in company value above a set baseline. Each addition would be at the board's discretion. The announcement does not disclose the baseline or valuation milestones.

Anthropic presents the valuation-linked equity allocation as an alternative to escalating executive compensation awards. That is the company's stated rationale for directing additional equity toward public benefit as its value grows.

The proposed share allocation also adds to pledges made by the company's founders and employees. Anthropic says its cofounders pledged at founding to donate 80% of their own equity holdings.

Approximately half of employees participate in an equity donation program that Anthropic matches, the company said Sunday. Anthropic puts the combined founder, employee and company commitments, including the new proposal, at roughly 20% of company shares. The new fund could receive additional shares at future valuation milestones, with each addition at the board's discretion.

Company ownership

Anthropic says it chose an LLC over a tax-exempt nonprofit or another structure to allow adjustments between different forms of giving. Large nonprofit donations and direct distributions are examples of possible uses, rather than programs announced for launch.

In a Summer 2018 analysis, Dana Brakman Reiser, a Brooklyn Law School professor, described philanthropic LLCs as offering flexibility, privacy and owner control compared with tax-exempt foundations.

LLC owners could retain broad discretion over spending and governance while avoiding disclosure obligations attached to traditional foundations, Reiser wrote. She described a cost to public accountability: the form gives the public less ability to examine, understand and influence philanthropic activities.

Know someone who'd find this useful? ✉️ Email it to a friend in one click, or they can subscribe free here.

Anthropic says none of its equity held by the LLC will be used for lobbying or political giving in the United States or abroad.

Details still to come

Its separate research fund's July 22 agenda includes tests of income support and ways for workers to hold stakes in AI-driven growth. The agenda envisages typical grants of $5 million to $30 million for institutional applicants, with no direct awards below $1 million. Eligible applicants include accredited universities and degree-granting institutions, independent research institutes and policy research organizations, and nonprofits with a track record of running field experiments at scale. Individual researchers can lead institutional proposals, but the research fund does not accept applications from researchers applying in a personal capacity. Those are research grant terms. They do not establish eligibility or payment amounts for the proposed Public Benefit Fund.

Sunday's release names no recipients or payout eligibility criteria and specifies no amount of cash available for distribution. It also gives no payout dates or timetable for turning the proposed share commitment into spending.

Anthropic says it will disclose more about the fund's structure and governance in the coming months, including the valuation milestones that could increase its stake. It also plans to publish a report on progress against its public benefit mission.

Frequently Asked Questions

Who could receive money, and when?

Nonprofits and specific populations could receive support, including efforts addressing AI's effects on workers. These are possible uses, with no programs announced for launch. The release names no recipients or payout eligibility criteria and specifies no amount of cash available for distribution. It also gives no payout dates or timetable for turning the proposed share commitment into spending.

How could the fund's share allocation grow?

The initial proposal would commit 5% of Anthropic's shares, subject to stockholder approval. The board could add another 2% of shares for every 50% increase in company value above a set baseline. Each addition would be discretionary. The baseline and valuation milestones have not been disclosed.

How do founder and employee pledges fit into the combined roughly 20% tally?

Cofounders pledged at founding to donate 80% of their own equity holdings. Approximately half of employees participate in an equity donation program that Anthropic matches. Anthropic puts the combined founder, employee and company commitments, including the new proposal, at roughly 20% of company shares.

Is this the same as the Economic Futures Research Fund?

No. The separate $200 million Economic Futures Research Fund supports external studies of responses to AI's economic effects. Its July 22 agenda includes income support tests and ways for workers to hold stakes in AI-driven growth. Its research grant terms do not establish eligibility or payment amounts for the proposed Public Benefit Fund.

Why use an LLC, and what does that mean for accountability?

Anthropic chose a wholly owned LLC to allow adjustments between forms of giving. In a Summer 2018 analysis of philanthropic LLCs, law professor Dana Brakman Reiser described flexibility, privacy and owner control, alongside reduced public ability to examine and influence philanthropic activities. That commentary predates this proposal. Anthropic plans to disclose more about the fund's structure and governance in the coming months.

AI-generated summary, reviewed by an editor. More on our AI guidelines.

Workers Say AI Took Over Colleagues' Tasks as Jobs Data Shows No Measurable Shift
Caroline Falkman Olsson co-authored a July Epoch AI report that split office work into ten activities, including reading documents, maintaining records and analyzing data, and asked who had done each
The Implicator
WSJ Survey Finds Top Economists Split Three Ways on AI Job Losses
Fifteen of the 16 economists The Wall Street Journal surveyed on AI and the future of work said the technology will meaningfully lift labor productivity, and none said it will not. The same panel, pub
The Implicator
Anthropic Files to Form AnthroPAC as AI Midterm Spending Tops $300 Million
Anthropic wants a seat at the table in Washington. The company filed an FEC statement of organization on Friday to launch AnthroPAC, a political action committee funded by employee donations, The Hill
The Implicator
AI News

San Francisco

Editor-in-Chief and founder of Implicator.ai. Former ARD correspondent and senior broadcast journalist with 10+ years covering tech. Writes daily briefings on policy and market developments. Based in San Francisco. E-mail: editor@implicator.ai