Anthropic Drafts Its Own AI Rulebook the Same Week Investors Mark It Down to a Mere $1.5 Trillion

Most companies preparing for a stock market debut spend the final weeks polishing a slide deck. Anthropic, the maker of the Claude chatbot, appears to have spent part of that time writing legislation.

POLITICO reported on Friday that Anthropic representatives have quietly pitched lawmakers on Capitol Hill a draft bill to regulate the safety and security of frontier AI, the most powerful models. Reuters followed up, and an Anthropic spokesperson confirmed that the bill reflects the company’s “Advanced AI Framework”, which it published earlier this year.

What the draft would do

According to the reports, the government could block a model’s release or use only if it posed a foreseeable, imminent risk of more than $1 billion in damages, or of killing or seriously injuring more than 50 people. The risks are limited to three kinds: a biological weapon, a model evading its developer’s controls, or a cyberattack.

Independent evaluators would register with the Commerce Department and periodically review AI labs and their frontier models. If they found an “imminent enumerated risk”, they would have to report it to Commerce within 72 hours. Commerce could then ask a federal judge in Washington, DC to suspend the model. The draft explicitly bars any other US department, agency or officer from halting development, deployment or internal use.

Anthropic’s spokesperson said the bill would give the government the “ability to block or deter dangerous deployments of the most powerful models”, while requiring developers to test their models, be transparent about the results and submit to independent evaluation.

Why this matters

There is a real case for this. Washington has spent years arguing about AI rules without passing many, and a company handing over concrete text at least gives lawmakers something specific to argue about.

But the details also show what a well-drafted bill can do for the company that drafted it. A single route to a pause, through one department and one court, is a lot more predictable than many state regulators each having a go. High thresholds mean most everyday harms fall outside the bill. Predictability is exactly what investors like to see in a prospectus.

Meanwhile, at the investor day

That brings us to the other Anthropic story of the week. Reuters reported that some current and prospective investors who attended the company’s investor day in San Francisco discussed a valuation of around $1.5 trillion for its November share sale, down from the roughly $2 trillion many had expected.

The reason is not that business is bad. Anthropic told investors that third-quarter revenue rose 45% from the previous quarter to $16.8 billion, and that it was profitable on an adjusted basis. The worry is that revenue had more than doubled the quarter before, to $11.6 billion. Reuters Breakingviews pointed to a price war on tokens, the units AI companies charge by, which lifted usage but cut revenue per unit.

The takeaway

Put together, the two stories describe a company trying to set the terms of its own future: how it can be stopped, and what it is worth. Neither is finished. The bill has not been introduced in Congress, and the IPO valuation is still being haggled over. Readers should watch both, and remember that “only $1.5 trillion” is a sentence that tells you how strange this market has become.

Commentary: opinion based on the reporting linked below. Nothing here is investment advice.