Why does Anthropic's IPO filing spend 80 pages on risk?

Everybody wants a piece of Anthropic. There is talk of a valuation up to $2 trillion, with the listing reportedly slipping to mid November. And then somebody actually reads the paperwork.

Reportedly, around 80 of roughly 261 pages are risk factors. That is more than the description of the business itself.

Two kinds of fear

The first is politics. Government contracts are under 1% of revenue, and yet the filing warns that the government's attitude could damage relationships with commercial customers and partners. It points to earlier clashes: an order for federal agencies to stop using the company's models, a supply chain risk label from the Pentagon, and export restrictions on its strongest models this summer.

The second one is bigger. The company writes that advanced AI could pose catastrophic, even existential, risks to humanity. In a document meant to sell shares. Right next to the stock details.

So is this honesty, or insurance? Is the company scared, or just covering itself so it can say "we warned you" later?

Probably both. Lawyers write exhaustive risk sections by nature, so nobody can accuse the company of hiding anything. But very few companies put a sentence about the fate of humanity next to a share price. Either way, one thing is clear: the risk follows them everywhere. Not just in the models and in what happens when people talk to them, but all the way to the moment they ring the bell at the exchange.