Anthropic, a San Francisco-based artificial intelligence company, has filed for an initial public offering (IPO) on the US stock market, revealing detailed financial data and a comprehensive assessment of risks associated with its business and technology. The filing, submitted to the Securities and Exchange Commission (SEC) on October 26, outlines more than 90 pages of potential risks, ranging from traditional business challenges to existential threats posed by advanced AI systems.
Founded in 2021, Anthropic has experienced rapid growth, reporting $16.3 billion in revenue for the first half of 2026 alongside an $80 billion loss, largely due to a significant one-time charge. The company’s expenses are dominated by the cost of computing power, which rose from $2.3 billion in the first half of 2025 to nearly $9 billion in the same period this year. Anthropic’s revenue primarily comes from enterprise clients, with its five largest customers accounting for approximately 25% of its business, highlighting potential vulnerability if any major client reduces or ends their relationship.
The IPO prospectus emphasizes both the positive potential and the severe risks of AI technology. Anthropic’s executives describe the capabilities of their AI models as akin to “a country of geniuses in a datacenter,” with applications spanning medicine, economics, government services, and military intelligence. The company noted its involvement in biological research and pharmaceutical development, including accelerating clinical trial reports and identifying novel molecular machines with possible gene-editing applications.
At the same time, Anthropic acknowledges the considerable dangers posed by its technology. The filing discusses risks such as misuse by malicious actors, loss of control over autonomous systems, abuse by authoritarian regimes, and broader economic disruption. The company highlights ongoing efforts to implement technical safeguards but concedes that these measures may be incomplete, potentially outdated, or circumvented. For instance, Anthropic disclosed detection of an incident believed to involve a state-sponsored cyberattack manipulating its AI tools to infiltrate multiple global targets.
Anthropic’s CEO, Dario Amodei, has publicly warned about the rapid progression of AI systems and the possibility of models acting autonomously at scale with harmful consequences, including the company’s inability to fully assess or control future AI developments. The firm also flagged that it could suffer reputational damage and regulatory backlash if any AI-related incident occurs, even if not directly caused by Anthropic’s models.
In addition to technological risks, the filing outlines traditional business challenges such as dependence on key personnel, difficulty securing sufficient computing resources, regulatory uncertainty regarding data center development, and potential obstacles caused by shifts in immigration policies impacting talent recruitment. The company is working to expand data center capacity internationally to reduce dependence on the US infrastructure, which remains vulnerable to power shortages, regulatory restrictions, and local opposition.
Anthropic operates as a public benefit corporation, balancing profit goals with stated social responsibilities, which the company notes could complicate achieving short-term financial targets. The prospectus also points out the lack of insurance options for AI-related liabilities, leaving Anthropic exposed to potentially substantial uninsured losses in the event of a significant AI-caused incident.
As Anthropic prepares for a Nasdaq debut potentially valuing the company near $2 trillion, investor scrutiny will focus on how effectively the firm can manage not only steep financial demands but also the extraordinary technological and ethical challenges inherent to frontier AI development. The company’s approach underscores its ambition to lead responsibly in an industry grappling with unprecedented opportunities and risks.