What a chat with Anthropic’s CEO told me about how AI may affect biotech

Source Domain: www.statnews.com

The most convincing thing Dario Amodei, the CEO and co-founder of Anthropic, said to me during an on-stage conversation last week was that perhaps his original vision of how AI would change biotech might not start to be visible for a decade.

In a 2024 essay, “Machines of Loving Grace,” Amodei had argued that artificial intelligence, and in particular large language models like Anthropic’s Claude, could allow researchers to make what we think of as a decade’s worth of progress every year, covering a century in a decade. Now he admits we’re not there yet.

“I don’t think that today we can make progress at a rate of ten years per year for a number of reasons,” Amodei said. Those included: Models aren’t as good as they someday will be; researchers need time to figure out how to use these tools; and the infrastructure and regulatory systems will take time to change.

Amodei and I were speaking at an Anthropic event where the company, a public-benefit corporation meant to be focused on improving the world, unveiled a product for biologists and pharmaceutical companies called Claude Science. I agreed to interview Amodei on-stage as part of the event, with the stipulation that I would decide on my own what to ask.

STAT+ Exclusive Story