When a company founded on the language of caution begins preparing for the stock market, people notice the tension. Anthropic, the maker of the Claude chatbot, has started laying the groundwork to become a publicly traded company, and the reaction has been as revealing as the news itself. On NPR, host Steve Inskeep put the question to Casey Newton, who hosts the podcast "Machine Gods" and has covered the technology industry for years. The tension is easy to describe.

Anthropic was created in 2021 by former OpenAI staff who said they wanted to build artificial intelligence more carefully, and it has long presented safety as a core part of its identity. A public listing, however, changes who the company answers to. Shareholders expect growth, regular financial reports and, in time, profit. Supporters argue that an offering is simply how a company that spends enormous sums on computing power raises the money it needs.

Sceptics ask whether a safety-first mission can survive quarterly earnings calls. The timing matters too. The race to build advanced AI is now financed on a scale once reserved for energy and telecommunications, with chips, data centres and electricity driving costs ever higher. In that setting, access to public capital markets is not a luxury but a strategic weapon, and rivals in the United States, China and Europe are watching closely.