Anthropic plans to set aside 5 percent of its shares for a public benefit fund

AI company Anthropic is creating the Anthropic Public Benefit Fund. The fund will direct a portion of the company's shares toward public benefit purposes. The plan is to set aside 5 percent of Anthropic's shares for the fund, but that requires shareholder approval. It has therefore not yet been adopted.

At its discretion, the board can add a further 2 percent for every 50 percent increase in the company's value above a set base value. The sources do not say what the base value is.

Flexible structure

The fund is being established as a wholly owned LLC, a US corporate form, and not as a nonprofit under section 501(c)(3). This is meant to provide flexibility in how the money can be used. The fund can make large donations to traditional nonprofit organizations or make direct payments to specific population groups. No specific payments have been disclosed.

The shares in the fund may not be used for lobbying or political contributions, either in the US or abroad.

Focus on jobs

Anthropic has already committed USD 200 million to research on the economic effects of AI on workers. According to Anthropic, the pace of AI development may mean that bolder measures are needed to address job disruption.

TRT World cites research from the International Monetary Fund, IMF, according to which AI could affect up to 60 percent of jobs in advanced economies. Jobs could be enhanced, diminished or eliminated.

According to TRT World, the cofounders have pledged 80 percent of their shares, and an employee program has participation from about half of the staff. Together with the fund, this amounts to roughly 20 percent of Anthropic's shares going to public benefit purposes. The employee portion depends on how many participate.

IPO and legislative proposal

According to Briefs.co, citing sources, formal marketing of an IPO could begin as early as the week of November 9. This does not appear in Anthropic's own text.

Briefs.co also reports that Anthropic has proposed a draft bill with safety thresholds to Congress. Under the draft, the government could block models that could lead to deaths or injuries to more than 50 people, or where the harm could exceed USD 1 billion.