When I was in engineering school years ago, I had this phrase "faith-based engineering" to describe when my roommate would show me a stress-strain analyses of various parts of the electric racing rig he was working on, we'd look at it, think vaguely about what we know/learned, and say fuck it, build it, see what happens. (no one died, we were in college).
edit: english is hard for english speaker without tea
My previous homeowner was also a vibe maintainer! The difference that I see in this is that the LLM is reasonably at 'expert' level for many of these things. If I sit down with Stevie Wonder and ask him to help me write a song, the resulting song is probably going to be pretty good. Stevie also knows by experience lots of things to avoid, and is intuitively going to help me avoid them unless I instruct him specifically to make a poor choice.
I think there are likely opportunities too to have models or system prompts that cater or adapt to the experience level of the person it's working with. "As you interact with the user, determine their relative level of knowledge and experience. If they seem to be relatively inexperienced with software development, be much more aggressive in helping to warn them about and avoid common pitfalls, bad architectural decisions, and security issues."
I suspect it's probably going to enable a lot of poor quality stuff, but it also may to some degree raise the floor of what's being produced at the same time.
I think the difference between electrical work and software work is with electrical a lot of those standards don’t really change from year to year outside of some code updates. Wiring a wall socket or moving a socket around has been pretty much solved for decades. maybe a different kind of outlet that is a little more safe or some exception about sockets near a sink.
Versus software I were there’s hundreds of different outlets hundreds of different wires tens of different storage mechanisms. Now if the LLM is even slightly unsure, it will hallucinate leading to a mess when things go wrong that the user doesn’t know how to fix
Then an actual expert will have to come in and try to understand what went wrong, which adds additional time than if it was just built right the first time
It's going to lower the floor, because lots of stuff that would otherwise not get produced will be made, and it will be hard to see how incompetent the people behind it are at first glance because the UI could be decent. Imagine putting your valuable information into one of these garbage apps only to find out that the app wasn't even written by a programmer! If it goes just a little farther, and the AI deploys it for you too, how many disclosure and data retention laws are going to be broken by it? You'd have to be awfully desperate to want to use one of these 100% AI monstrosities created by an amateur.
Is it "artisanal" if almost every application still has to be made by humans? I think that nobody will want to run amateur AI software projects from strangers, unless it can be done in a sandbox. Even a well-meaning "creator" could have malicious garbage injected into their product by AI.
This is no different than what amateurs have been doing for a long time. Hopefully they’re not taking in PII or charging anyone for these apps.