Anthropic CEO Says IPO Will Not Change Its AI Safety Pace
Dario Amodei said the company will set training and deployment timelines according to the time needed to gain confidence that its systems are safe and remain under human control.

Anthropic CEO Dario Amodei said the company will set the pace of AI training and deployment according to the time needed to gain confidence that its systems are safe and remain under human control, despite commercial or political pressure.
Amodei said Anthropic was founded on the belief that powerful technology should serve human welfare. He pointed to possible advances in science and medicine, while saying results so far are early and experiments and clinical trials still take time. AI could accelerate those processes, he said.
He also identified risks from individuals or governments misusing AI, economic disruption and losing control of AI systems. Those risks are avoidable, Amodei said, and depend on choices made by people building AI.
Anthropic’s safety work includes protections against biological and cyber risks, alignment training and testing, and efforts to examine how models work internally, Amodei said.
He described a strategy of building a successful company while making safety-conscious choices, with the aim of improving the industry even when Anthropic is not leading. Amodei said the IPO strengthened the company’s finances and its resolve to do the right thing.