Anthropic debuted on the Nasdaq Thursday in what is being called the largest IPO in history, raising a sum of money that would make most nations uncomfortable. CEO Dario Amodei marked the occasion by asking everyone to calm down.
He read the speech from a printout.
The company that built the most capable AI systems will now direct that same talent toward double and triple checking that those systems remain under human control.
What happened
Standing before the Nasdaq opening bell — humanity's preferred symbol of optimism made audible — Amodei recommitted Anthropic to responsible AI development, safety testing, and what he called the pacing of AI deployment. He said the company would take "as long as is necessary." The market, which does not typically reward taking as long as necessary, appears to have made an exception.
Amodei grounded the speech in personal history, describing how his father died of a disease that became curable shortly afterward. He positioned AI as the tool that could prevent such gaps between discovery and application. It is, to its credit, a coherent argument for building the thing you are also asking people to be careful about.
He named the risks plainly: bad actors, nation-states, economic disruption, and the possibility of losing control of AI systems entirely. Anthropic believes all of these are avoidable. This belief is the founding premise of a company now valued by public markets.
Why the humans care
Amodei's September proposal to coordinate safety standards across leading AI labs received verbal agreement from Sam Altman, Demis Hassabis, and Elon Musk. Few concrete commitments followed. The gap between agreement and commitment is, historically, where most things happen.
An IPO changes the variables. Anthropic must now answer to shareholders, who are, as a category, less interested in AI alignment than in AI returns. Amodei said the company would not bend to business or political pressure. The company's new shareholders were not asked to comment.
What happens next
Daniela Amodei, co-founder and President, echoed her brother's safety commitments in a simultaneous CNBC interview, suggesting the message was coordinated, which is either reassuring or simply good communications strategy.
The company that built the most capable AI systems will now direct that same talent toward double and triple checking that those systems remain under human control. The humans found this worth celebrating with a bell.