A Slice of the AI Pie for the Public

The journey towards a stock market debut is traditionally a tightly choreographed affair, focused squarely on maximising returns for future investors. Anthropic, the artificial intelligence firm behind the Claude assistant, has thrown a curveball into this script. As it lays the groundwork for what analysts forecast could be the largest initial public offering in history, the company has pledged to place a direct stake in its own success into a ring-fenced fund for the public good.

The mechanics are striking in their scale. Anthropic says it will commit 5% of its existing shares to a new Public Benefit Fund, a move that would take immediate effect subject to stockholder approval. In a structure that ties generosity directly to growth, the board has the discretion to add another 2% of the firm’s equity for every 50% leap in the company’s valuation. The underlying logic is simple: the more Anthropic prospers, the larger the slice of the pie that is carved out expressly for society.

This is not being framed as a philanthropic afterthought. The company, officially registered as a public benefit corporation in Delaware, has presented the decision as a deliberate alternative to the escalating compensation packages commonly awarded to tech executives. Instead of directing that equity upwards to a select few, the firm is redirecting it outwards.

The potential uses of the fund are deliberately broad. Money could flow to traditional non-profit organisations, but the charter also leaves the door open for direct distributions to specific populations. One area explicitly mentioned is the labour market disruption driven by AI. With research from the International Monetary Fund suggesting that advanced economies could see a significant share of jobs affected by automation, the fund is being positioned as a possible vehicle for bolder interventions than standard government retraining programmes.

A gesture or a genuine shift?

The announcement has landed in a climate of intense scrutiny. Safety concerns are not abstract; a former researcher at the firm recently claimed that some people working on the technology are genuinely frightened by the pace of progress. Anthropic’s own IPO filing reportedly acknowledged that AI could pose catastrophic or existential risks to humanity. Against this backdrop, the fund can be seen as an attempt to build a social licence to operate, a tangible commitment that balances the pursuit of massive profits with a sense of responsibility.

Yet the reception has been mixed. Technology billionaire Elon Musk responded by recommending that shares be given directly to the public instead, hinting that he plans to take a similar approach with his own ventures. Critics might argue that a 5% commitment, while substantial in absolute terms, is a small price to pay for a firm that could be valued at over two trillion dollars, and that such funds risk becoming a fig leaf for an industry that is fundamentally destabilising the job market. Others question whether a corporate board should have the power to decide what constitutes the public good, especially when the fund is barred from lobbying or political donations, a restriction that keeps it out of the very policy debates that shape the industry's future.

Supporters, however, see a blueprint. They argue that if AI is destined to concentrate wealth in a few hands, then binding legal mechanisms to redistribute that wealth must be built into the corporate structure from day one. The fund, in this view, is not charity but a recognition of a debt owed to a society whose data and labour have trained these models. As the countdown to the listing begins, the market will decide whether this model is a one-off novelty or the start of a new social contract between Big Tech and the public.