How AI is Rewriting the Code at Top AI Labs

Hey there—so I stumbled across a wild claim on Reddit the other day. Turns out, companies like Anthropic and OpenAI are now letting AI write all their code. Yep, 100%. That means their engineers aren’t typing a single line manually anymore. Crazy, right?

The post links to a Fortune article (and Reddit thread) where developers from these companies talk about how AI tools handle everything from simple scripts to complex systems. They’re using models like their own to automate debugging, testing, and even designing new features. It’s not just about saving time—it’s about shifting their focus to bigger-picture problems.

I get that this sounds like pure sci-fi, but it’s real. And honestly? It’s kind of impressive. Imagine trusting AI to build the backbone of your AI tools. It’s like a robot building its own factory floor. But there’s some nuance here: these companies still have human engineers overseeing the process. It’s not “no humans involved”—it’s more like “humans in the loop, but with hands off the keyboard.”

What does this mean for the rest of us? Well, if AI can handle code this well, we might see a flood of new tools that make programming way more accessible. Or maybe we’ll all just get better at asking the right questions. Either way, it’s a glimpse into what the future of work could look like.

Want to dig deeper? Check out the Fortune article linked in the Reddit post. And if you’ve used AI for coding before, I’d love to hear your thoughts—does it actually work as well as they say?