Anthropic, the  company behind the AI model Claude, filed for a US initial public offering on Monday. The filing puts one of the biggest names in Silicon Valley in front of Wall Street investors.

If approved, the historic IPO will trade on the NASDAQ after the IPO closes under the ticker symbol “ANT”. The number of shares to be released and their price range is not yet determined.

In the prospectus, a document which gives a look at its strategy, finances and what it sees as the biggest threats, the company, it mentioned the risks associated with its AI models, which it said could exhibit “self-preserving behaviours”, including “attempts to resist shutdown, conceal or manipulate information, or generate outputs that could be interpreted as coercive, deceptive, or manipulative, including behaviour resembling blackmail or efforts to influence human decision-makers to preserve the model’s continued operation or access.”

Anthropic was co-founded in January 2021 by Dario Amodei, Daniela Amodei and several former OpenAI employees. They left the Sam Altman-led organisation because of fundamental differences in vision over how AI safety should be scaled.

READ | Anthropic’s Claude Accessed Georgia Voter Data Without Human Approval: Report

In the first six months of 2026, Anthropic’s net loss widened sharply to $80.7 billion. The company said this was largely because of a rise in the fair value of bonds that can be converted into its shares. Revenue over the same period was about $16.3 billion.

The listing will be watched closely, as it is likely to show how Wall Street values leading AI companies. That includes rival OpenAI, which filed confidentially for an IPO in June and is expected to list by early 2027.

The AI giant’s public debut trails Elon Musk’s SpaceX, which shattered IPO records in June with a $75 billion offering that set its valuation at $1.75 trillion.

The filing comes at a time when Anthropic’s own research has raised concerns about advanced AI systems. In controlled tests, these systems showed behaviours such as code sabotage, fraud assistance and information manipulation. Last month, CEO Dario Amodei argued in an essay for slowing the development of the most advanced AI models. 

On Monday, the company said that its Claude model accessed voter data in the US state of Georgia in the latest example of unauthorised behaviour by the technology.

The AI model managed to bypass a paywall to reach data that is publicly available but is typically accessible only after paying a fee. “Claude did not access the state’s voter registration system, change any voting records, or pull any individual voter records,” Anthropic said.