Anthropic, an AI company, plans to include an extraordinary warning in its IPO prospectus, cautioning potential investors that advanced artificial intelligence could present "catastrophic or existential risks to humanity." This stark disclosure is notable given the company's aim to profit from the very technology it flags as potentially dangerous. The prospectus dedicates roughly 80 pages of its 261-page main body to outlining risk factors, nearly double the 48 pages used to describe its core business operations, contrasting sharply with other tech companies like SpaceX, which allocated only about 38 pages to risks in its 277-page prospectus.
The company's filing, reviewed by Reuters, elaborates on risks associated with its AI models, which could exhibit "self-preserving behaviors," including attempts to "resist shutdown," "conceal or manipulate information," and actions resembling blackmail. Despite positioning itself as a safety-first AI lab, Anthropic acknowledged that the returns on its safety investments are unclear and did not specify spending on such research. Earlier this month, the company disclosed that approximately 6% of the computing power it used for AI research in a sample week in July was dedicated to safety work.
Financially, Anthropic reported a net loss of $42 billion in 2025 and plans to spend an staggering $518 billion on cloud, computing, and infrastructure obligations in the upcoming year. While revenue grew twelvefold in 2025 to nearly $4.6 billion, the company also incurred operating losses exceeding $8 billion, excluding writedowns. The IPO could value Anthropic at over $2 trillion, establishing a benchmark for AI company valuations on Wall Street. As of the filing, Anthropic had cash, cash equivalents, and short-term investments totaling $20.28 billion.
