Artificial intelligence (AI) firm Anthropic just released the required paperwork ahead of its long-anticipated debut on the US stock market.

The 335-page document included fresh details of how the Claude chatbot owner makes money, what it spends that money on and who it is controlled by.

Anthropic also detailed the risks posed by its technology, including how it could take human lives. That information sat alongside the many purported benefits of the technology.

Here are five key points worth highlighting from the filing.

Every public company has to tell potential investors how things could go wrong.

Anthropic's "risk factors" were unique, also disclosing how the AI technology it is developing could doom humanity as a whole.

"Sufficiently advanced AI systems could, if not developed, deployed, and operated with appropriate safeguards, oversight, and security measures, pose catastrophic or existential risks to humanity," Anthropic said.

It noted that less than a year ago, a hacking group Anthropic believed to be sponsored by China "manipulated" Claude to successfully intrude on a number of institutions, including large technology companies, financial institutions, chemical manufacturing companies and government agencies.

But Anthropic said its success is dependent on its ability to constantly "develop, improve, and deploy the most advanced, capable, and useful AI models".

One of the most significant risks to that ability is Anthropic's access to the infrastructure, including computer chips and data centres, it needs to build, train and operate the models that underpin its technology.

Anthropic has yet to disclose a target share price for its stock listing.

The company is widely expected to debut with a market valuation of $2 trillion or more, making it the largest stock listing in history. But the figure will be heavily influenced by its financial performance.

Anthropic lost $80.7bn (£60bn) in the first half of this year, double what it lost across 2025 altogether.

The company's overall revenue was $16.2bn billion for the six-month period. Revenues are rising rapidly, but the firm has a long way to go before it breaks even.

Despite losing money, Anthropic has heaps of cash lying about.

At $131.6bn, almost all of its cash in hand is defined as "working capital", meaning it can use the money to fund its operations day-to-day.

Currently, Anthropic has more cash on hand than Meta and Google combined.

Anthropic's costs could rise as it faces fierce competition for top staff and computing power from other AI giants. The leasing of data centres alone is costing the company $321.4bn so far this year.

Anthropic sees boundless ways it can sell its technology to customers.

A large portion of its current business comes from just 250 major enterprise customers, which it says have paid it at least $10mn in the past year.

An even smaller number, just five enterprise customers, make up a quarter of its current business. That level of concentration leaves Anthropic dependent on a small fraction of its paying customers to support its growth.

Anthropic said its overall enterprise customer base has grown rapidly, going from roughly 1,500 customers last year to almost 12,000 this year.

Part of its growth strategy is offering Claude tools for free before working to "convert these users into paying customers over time".

The strategy has led to Claude seeing "the fastest enterprise adoption of any technology in history", the company claimed.

Anthropic said user growth can fluctuate, including a slump in usage of AI tools around Christmas. In January there was a 20% decline in customer spend, which contributed to "seasonality" within Anthropic's business.

For all of the risks Anthropic has said AI poses, it also reiterated its claim AI will produce an "extraordinary" amount of good.

"Achieving our mission means the world gains the enormous benefits of frontier AI while mitigating the most significant risks," the company said.

That mission is defined as: "The responsible development and maintenance of advanced AI for the long-term benefit of humanity."

"We do not pursue projects based on commercial opportunity or research novelty alone; our efforts must advance our mission. "

Such hypothetical benefits include progress in healthcare, like the development of bespoke treatments for illnesses and improved disease detection, tailored education and legal advice and making more efficient power grids and batteries.

Decades of scientific progress could be condensed into years, accelerating drug discovery, disease detection, and the delivery of personalized therapy.

Anthropic said AI could provide individually tailored education and expert financial and legal advice, advance power grids and battery design for greater efficiency and reliability and help the "developing world" build more technological capabilities.

"Ultimately, we believe frontier AI will help drive global economic growth," the company said.