Anthropic files for IPO, revealing explosive growth and AI risks

Anthropic has publicly released its IPO filing, confirming explosive revenue growth for the Claude chatbot creator while detailing significant financial losses and extensive AI-related risks. The San Francisco company reported $16.3 billion in first-half revenue, a dramatic increase from $4.6 billion for all of 2025. The filing shows Anthropic remained deeply unprofitable, with losses of $8.4 billion even after excluding a $72 billion accounting effect from stock valuation. The company has committed $265 billion over 20 years to cloud computing providers for AI development capacity, and investors have reportedly pushed to lower its valuation from $2,000 billion to $1,500 billion before a possible November listing. Anthropic devotes nearly 100 of 335 pages to risk factors, warning that its safety guardrails have already failed or been misconfigured, and that AI models could cause damage through unauthorized actions. The company also highlights risks from recursive self-improvement, where flaws could propagate through future models, and reserves the right to delay launches, as it did with the Mythos model in April.