Soundtrack: Mack Glocky - Chasing Cars
Last week, I spent a great deal of time and words framing the generative AI industry as a cynical con where OpenAI's Sam Altman and Anthropic's Dario Amodei have used a compliant media and braindead investors to frame unprofitable, unsustainable, environmentally-damaging and mediocre cloud
My case in point. Instead of farming out translation jobs for 12 languages in WPML to people, we now auto translate all content via AI. That’s jobs lost. Complicated legal docs is an edge case in the content world.
Cursor is agent mode with Claude 3.7 absolutely writes well enough after indexing our codebase to mean less junior devs. I’m sorry that doesn’t fit into your narrative.
That’s not to say the world isn’t full of AI slop, and there aren’t huge issues around LLMs, But there are some solid use cases for this technology in replacing meat bags.
I tried cursor, claude, copilot. They’re not good.
Like, they can sometimes generate 10 lines of perfectly reasonable code, but they constantly completely misunderstand my intention or simply produce garbage. But the garbage looks just good enough, that you actually have to read and understand it, which slows me down.
Maybe we’re operating in completely different worlds, but even for the junior devs in my company, typing speed was never an issue. I’m sure, within a few years LLMs can generate much better code, but I don’t see widespread unemployment. They need way too much babysitting and result in worse code.