Silicon Valley’s plan to automate everything
Illustration by The Atlantic

Silicon Valley’s plan to automate everything

Over the past few weeks, OpenAI, Anthropic, Google, and other major AI companies have launched products that are getting better at doing the tasks usually assigned to humans. Claude Code, a new coding program from Anthropic, can complete much of a software developer’s job—but far faster. Google added a widely available “workhorse model” for daily tasks. “Humans will still nudge these models along, of course, but they are engineered to help fewer people do the work of many,” Atlantic staff writer Matteo Wong writes.

Exactly how AI will be implemented in different workplaces is still unclear, but the technology’s advances are already creeping into the highest levels of governance. The Trump administration is reportedly testing a new chatbot with 1,500 federal employees at the General Services Administration and may release it to the entire agency soon—meaning that generative AI could be used to automate work by more than 10,000 workers who are responsible for more than $100 billion in contracts and services, according to Matteo. “Of course, federal agencies have been experimenting with generative AI for many months,” he writes. But given how much refinement this technology still needs, implementing its use carries important risks—especially if it’s not subjected to thorough testing, strict guardrails, and public transparency.

Today’s newsletter brings you stories about how AI can change the workforce:

Stephanie Bai, associate editor

Erik van der Kooij

Founding director Feeling Europe FNDN, founder @C21fellow. Connected to NEXUS Institute | CEPS | Caux Round Table for Moral Capitalism | Council for Inclusive Capitalism | CIFA | Odyssey | GBBC

1 天前
回复
Carl Williams

Senior All-Source Analyst | MGM, MBA, MAF

1 天前

And fire everyone.

This ought to be a matter of great concern. But instead we are facing a disabling technocracy.

要查看或添加评论,请登录

The Atlantic的更多文章