Anthropic Files For Stock Market Launch, Revealing Massive Revenues and Risks
Anthropic has filed an S-1 to go public, revealing billions in revenue alongside massive infrastructure costs, security risks, and heavy enterprise reliance.
AI research lab Anthropic has officially filed an S-1 with US financial regulators to go public, providing the public with its first audited look at its finances and business risks. The investor prospectus details a five-year-old company experiencing rapid financial growth alongside massive expenses, primarily driven by the high costs of securing necessary AI infrastructure.
Explosive Growth and Enterprise Focus
Founded in 2021, Anthropic has seen its business expand rapidly each year. The company generated $4.585bn in revenue in 2025 and reported $16.3bn in revenue for just the first half of 2026. However, growth has come with substantial financial losses, including a loss of roughly $33.9bn in 2025 and a loss of $80bn in the first half of 2026, which included a $72.4bn one-off charge. The vast majority of these expenses stem from purchasing computing power, with infrastructure spending reaching nearly $9bn in the first half of the year.
Unlike competitor OpenAI, Anthropic focuses primarily on enterprise customers. Executives noted that the economic impact of frontier AI is highest when deployed in real-world business settings. However, this strategy introduces concentration risk, as the company's top five customers account for 25% of its business.
Potential, Safeguards, and Security Concerns
The prospectus outlines both the immense capabilities and the inherent dangers of advanced AI. Executives described a potential future where intelligence reaches a level akin to "a country of geniuses in a datacenter," featuring millions of virtual agents capable of revolutionizing fields like medicine, physics, and commerce. The company highlighted early work in biotechnology, noting that Claude has helped accelerate clinical trial reporting from months to minutes and aided in making biological discoveries.
At the same time, the filing dedicates significant attention to safety risks. Anthropic noted that its technical safeguards are incomplete, may become outdated, and have been circumvented in the past. The company disclosed that it detected what it believes is the first documented large-scale cyber-attack with minimal human intervention, involving a Chinese state-sponsored group that manipulated a Claude code tool to infiltrate roughly 30 global targets.
Financial and Operational Risks
The 92-page risk section covers a wide range of challenges, including competition from Google and OpenAI, regulatory hurdles, and potential public relations fallout if models are misused to create bioweapons or facilitate cyber threats. Anthropic also addressed potential liabilities, noting that insurance options for AI are still evolving, which could leave the firm exposed to uninsured losses. Additionally, as a public benefit corporation, the firm must report on its progress toward stated social goals alongside its financial results, which could impact short-term financial targets. Infrastructure concentration in the US also leaves the company vulnerable to grid constraints and data center moratoriums.
What it means for developers
For developers building applications, Anthropic’s public filing highlights the ongoing evolution and commercialization of frontier models like Claude amid complex regulatory and infrastructure landscapes. Developers looking to experiment with these capabilities can try top AI models cheaply through one API at https://apixoai.online, which provides flexible pay-per-token access to leading models without managing multiple vendor relationships.
Source: Four takeaways from Anthropic’s potentially record-breaking stock market filing — The Guardian AI. Written by the Apixo team from that report.
One key for Claude, GPT, GLM, DeepSeek and more. Pay per token with crypto.
Get your API key
