@Jonathanglick @TedUnderwood @brendannyhan It would be kind of funny if the only economically significant outcome of this for the next 10-15 years was largely as a substitute for human programmers (rather than the writers and chattering class as people keep predicting)

@adamgurri @Jonathanglick @TedUnderwood @brendannyhan As a software engineer who has fully incorporated AI into my workflow, I can speak to that:

Maybe it will change in the future, but for now, the AI's inability to reason holds it back. It's great at certain smaller self-contained tasks (especially visual scaffolding and formulaic code) that are typically the domain of low-level coders, but the high-level architecture of major programs, connecting the pieces, and overall flow (things like UX), are still very much in the realm of humans.

I do ultimately think AI will automate away some programming jobs, similar to how WordPress shifted some jobs from developers to marketers, but like most machines in an expressive medium, it's much better at quantity than quality.

That said, however, the AI has a far different function that people may not expect: It's great at teaching and reinforcing patterns to human programmers. That could be huge down the line. But right now, the patterns it teaches are... let's say, not always the greatest (because of what it was trained on). So you still need someone experienced to parse out the good from the bad. For inexperienced programmers, AI could just as easily reinforce bad habits as good.

Sign in to participate in the conversation
Qoto Mastodon

QOTO: Question Others to Teach Ourselves
An inclusive, Academic Freedom, instance
All cultures welcome.
Hate speech and harassment strictly forbidden.