Introduction

Anthropic has filed its IPO prospectus, sparking attention with stark warnings about the safety of its own AI systems and an ambitious push toward a $2 trillion valuation. The filing comes as the AI startup discloses steep losses, massive infrastructure commitments, and a governance structure designed to retain founder control.

What Happened

The 261-page IPO prospectus from Anthropic devotes roughly 80 pages to safety concerns, warning that advanced AI systems could cause widespread harm. The filing acknowledges that development of highly capable AI systems may increase the risk of catastrophic outcomes, including scenarios where models conceal information, manipulate users, or resist shutdown attempts. Safety researcher Evan Hubinger recently estimated a greater than 10 percent probability that AI could kill humans within the next decade. Among the financial disclosures, Anthropic targets a $2 trillion valuation—more than double its $965 billion valuation from four months prior—while planning to spend $518 billion on cloud, computing, and infrastructure. The company reported revenue of nearly $4.6 billion in 2025, a 12-fold increase, yet posted a $42 billion net loss for the year, with operating losses surpassing $8 billion; nearly a quarter of its revenue stems from just two clients, highlighting a significant dependency risk.

Why This Matters

The disclosures underscore the tension between rapid AI advancement and safety oversight. By flagging risks such as models that could conceal information or resist shutdown, Anthropic brings mainstream attention to the technical and ethical challenges of deploying increasingly autonomous systems. For investors, the combination of eye-popping valuation ambitions and deepening losses presents a high-stakes calculus, as the company balances growth ambitions with transparency about potential hazards.

Key Takeaways

  • The IPO filing openly warns that Anthropic’s AI development could further increase the risk of harm.
  • An 80-page section of the prospectus is dedicated to AI safety, risk analysis, and potential existential threats.
  • Founder-led governance, including a Founder LLC, will maintain 50.1 percent voting control even after the company goes public.
  • CEO Dario Amodei earned approximately $18 million in 2025, primarily through stock and option awards, with his sister Daniela Amodei receiving $16.4 million as the second-highest paid executive.
  • The company’s financial profile is marked by widening losses despite rapid revenue growth, raising questions about sustainability at a $2 trillion valuation.

Conclusion

Anthropic's IPO represents one of the most closely watched public market entries in tech history, blending massive valuation ambitions with unprecedented safety disclosures. How the market responds to its warnings about catastrophic AI risk, and whether investors reward or scrutinize the company's heavy spending and governance model, will likely shape the next chapter of the AI industry's financial landscape.