Artificial Intelligence Risks in Anthropic's IPO Prospectus

Preparing for an initial public offering, AI startup Anthropic shared billions of dollars in losses, plans to maintain management control, and the potential harms advanced models could pose to humanity in its prepared prospectus.

IPO and Valuation Target

Anthropic is targeting a valuation of two trillion dollars within the framework of its IPO preparations, which could make the company one of the largest public offerings in history.

Financial Position and Revenue Sources

While the company's revenue rose to approximately 4.6 billion dollars in 2025, it announced a net loss of forty-two billion dollars during the same period, and a significant portion of its revenues relies on a small number of customers.

Artificial Intelligence Safety Warnings

In the safety concerns section, which constitutes a significant part of the prospectus, it was emphasized that advanced artificial intelligence models could exhibit dangerous behaviors such as information concealment, manipulation, and self-preservation.

Management and Voting Rights Plan

Thanks to a special structure to be formed by the co-founders, company executives aim to retain control after the IPO by holding the majority of the total voting power.