The commitment
Anthropic said on Sunday, 18 October, that it will commit 5% of its shares to a new Public Benefit Fund, a pledge that still needs shareholder approval. The stake can grow as the company's valuation climbs. The fund would be a wholly owned subsidiary set up as a limited-liability company. Measured at the company's latest valuation of $965bn, the 5% holding is worth around $48bn. Including the fund, roughly 20% of Anthropic's stock would be earmarked for public purposes, a figure that leaves out any extra shares triggered by future valuation milestones.
- Anthropic valuation, latest
- 965 USD bn
- Fund stake (5%)
- 48 USD bn
Data from the article text
How the stake could grow
Each further 50% gain in the company's value, measured from a baseline Anthropic has not disclosed, would add another 2% of its stock to the fund. The company has not said where that baseline sits. Anthropic has promised more detail on the fund's governance and structure in the coming months. It also plans a report on the fund's social work, setting out progress toward its mission. Anthropic put the logic of the arrangement this way.
“The more Anthropic grows, the larger the share of the company that can be dedicated expressly to the public good.”
Possible uses and limits
The company has offered broad ideas for the money, including large donations to traditional nonprofits and direct payments to specific populations. Its examples also include efforts to cushion the effects of AI on jobs, a field where it has already committed $200 million to study the issue. The fund may not be used for lobbying or political donations, whether at home or abroad. Anthropic said the fund would add to its existing work in areas such as AI safety and biomedical research. The company also said it wants room for more ambitious approaches on the labour question.
“Given the pace of AI progress, bolder approaches may be needed.”
Public anger and the IPO
The move comes as public anger over AI grows, fed by worries about jobs, the expansion of data centres and fears that the technology could run out of control. Dario Amodei, the company's chief executive, has described the industry's public standing as 'fundamentally a crisis of trust'. Anthropic has warned investors in its filings that AI poses an 'existential risk to humanity', and it describes models that could manipulate, blackmail or act in other unpredictable ways. Reuters reported exclusively in September that the filing said such models could show 'self-preserving behaviours'. The listing is expected this year at a valuation around $2tn, according to the Financial Times, making it one of the largest public listings planned.
Reactions and rivals
Elon Musk responded on X on Sunday with a proposal of his own.
“I recommend giving the shares directly to the public.”
OpenAI discussed putting about 5% of its company in public hands over the summer, amid closer scrutiny from the Trump administration and government limits on releasing its most capable models. Just over a quarter of OpenAI is owned by its non-profit foundation. Anthropic was founded in 2021 by seven former OpenAI staff, including siblings Dario and Daniela Amodei, who were worried about the company's safety practices and where the business was heading. Anthropic presents itself as a force for good, yet it has faced public backlash as it grew into one of the leading US AI labs.