Urgent.News

What's breaking now, across hundreds of outlets.

AI

Can Anthropic afford to put safety first?

Today, I’m talking with Hayden Field, The Verge’s senior AI reporter, about the Anthropic IPO for the third and final episode in our series about AI in real life. The AI lab behind Claude went public on the Nasdaq last Thursday. The numbers are gigantic, but not really that important for this conversation, except to […]

Can Anthropic afford to put safety first?

Today, The Verge's senior AI reporter Hayden Field joins to discuss Anthropic's IPO and the challenges of balancing AI safety and profitability. Anthropic, the creator of the popular AI chatbot Claude, went public on the Nasdaq last Thursday, making it the first pure AI lab to do so. The company's valuation is enormous, but its real value lies in its ability to continue investing in safety while also satisfying shareholders and funding growing compute costs.

The IPO generated significant buzz, with new millionaires and billionaires emerging from the deal. However, the focus on safety branding now puts Anthropic in direct conflict with its financial interests. OpenAI, on the other hand, seems content to let Anthropic pave the way, as they have faced their own share of scandals and negative press.

Rewritten by urgent.news; may contain errors — check the original.

Read the original at theverge.com →

Report a problem with this story

More in AI

6 Delphi behaviors that silently break when an AI ports your code to C#, Java or C++

AI agents are good at producing code that compiles. With Delphi, "compiles" and "behaves the same" are far apart. Here are six behaviors I keep seeing agents get wrong — each one is measured on the…

  • Delphi's Round function behaves differently in C#, Java, and C++
  • Delphi allows unit replacement of built-in functions, causing unexpected behavior
  • Delphi strings are 1-based, while C# and Java use 0-based indexing

More from Saturday 21 November →