I think a lot of the things that separate successful/productive people from non isn't skill it's desire.

Why don't people make their own board games? Because it's not the programming part that's hard, it's the game design part, and some people have a creative impulse to do things and some people don't. A big part of the reason people don't becomes programmers isn't because they don't want to memorize all that weird syntax, it's because they don't want to sit inside a room, staring at a computer, writing programs.

Within reason (there are some counter examples to this) but having to develop a skill to accomplish something can be a good differentiator into the kinds of people who will have the passion to stick with something and the kinds of people that don't. Sitting down to learn Python is the easy part, if you don't have the patience to learn python, you probably don't have the patience for game design.

They all became programmers or adjacent. You and I could also learn any number of hobbies or start writing books with AI (if you really want to use AI, though there are tons of resources on the internet for how to do a lot of these kinds of hobbies or projects without AI). The ability to do something doesn't imply the desire to do it, and vibe coding still requires a technical desire that a lot of people self selected out of.

old man hyperbole: it's because skill and creativity are mutually co-dependent. It takes a game developer to invent UNO from a stack of playing cards. A man without (some) skills in a field cannot be creative in the field.

You need to describe to AI what you want quite precisely. Non-expert people can't describe basic things, let alone complex projects.

Because it is a narrative scam. Giving the illusion that non-technical people are also software engineers / programmers. (Which they are not)

Vibe coding just helps those who are already technical but minus the tests, TDD, or architecture in software engineering which is completely against clean code principles and they don't care if the code is correct or not.

It's slopware engineering whilst celebrating technical debt.

An LLM can give you working code if you're lucky but putting it to work takes some skill and probably the code it gives you doesn't quite work right and you need the skill and insight to figure out why and fix it yourself or give a detailed explanation of what it has to do.

I skipped the regular CS requirement as an undergrad because I'd been coding since I was 9 in BASIC and Assembly and Turbo Pascal and C and it would have been silly to take another Pascal course like the one I had in high school. I wound up taking 1-credit courses in C and TeX instead from John Shipman [1] [2]

There was a time though when I was helping out a friend with CS 101 and we decided to try to steal somebody else's code. Well, finding a home directory which had been

  chmod o+rx 'ed

was not hard at all but the program that we stole didn't work properly and we had to spend time debugging and fixing it, not to mention changing things around a little so it wouldn't be obvious we stole it. Looking back though, I see this was better preparation for being a professional programmer since a professional programmer spends a lot of time fixing bugs in programs other people wrote. Vibe coding has plenty of that.

[1] https://www.nmt.edu/news/2022/shipman-honorary-doctorate.php

[2] 1-credit classes with 3 credits worth of homework

Most people are consumers, not creators.

Aspects of coding not the primary focus for non-programers. aka unless need/require 'new' coding method(s), 'old'/know/paid-for stuff works just as well. Similar concept to, can not find someone in 'coding' field that's offering service, so "non-coder" comes up with something to fit situation. aka unix, George Lucas' Industrial Light and Magic, python, matlab, etc.

For areas 'not focused exclusvely/around coding', takes a bit more time for things (experience, rules, resources, training, how 'new' fits in) to move into main stream. Libral Arts coding perspective : https://medium.com/@markguzdial/learning-to-program-matters-...

'time' tech example (concept: why can't AI cameras serve as traditional punch clock) : https://tcpsoftware.com/blog/time-clocks/

Historical -> requirement to take a class in formal logic as part of undergraduate degree (vs. more degree specific formal logic)

For the same reason that most people who aren't writers aren't interested in vibe writing or the same reason why most non artists aren't interested in vibe art creation.

How many novels have you written using AI?

How many art shows have you had from your ai generated art?

People ignore vibe coding for the same reason they aren't vibe writing or vibe art creation, because either they don't know they can do these things or they just have other interests.

My hypothesis is that the average person does not really need or want more software. Decent consumer-facing software exists for virtually any use-case you can think of. The bottleneck is free time and attention not lack of software.

Nobody has yet even heard of vibe coding. It only happened yesterday.

Education and programmability. Schools have been chipping (extremely slowly, FAR too slowly) at the power of programmability of things. But where are we now? Few people program their calculator after school requirements. Few people write little programs for day to day tasks. More people do use spreadsheets (I expect with very minimal programming). For one thing, it seems programming is kept in math and programming classes, rather than use spreadsheets here and there in all classes?

But games? I don't see why non-programmers should want to make their own games. Good enough games are free to cheap - so why games? Games would be then just from the urge to create. I expect more people might want to create or alter clothes or create drawings, than games. Certainly no reason for "most people" to head into games development.

>You would expect most non-programmers to at least want to make their own games.

"non-programmers" comprising the entirety of the human race not currently engaged in computer science as a hobby or profession. No, most non-programmers don't want to make their own games. Most people on the planet have other interests and concerns.

Some "non-programmers" do want to make their own games, and they learn to use one of the many available game frameworks, and sometimes they even succeed.

I think the set of people passionate enough to want to make their own games, but also lazy enough to refuse to put in even the minimal effort necessary to learn how to do so, is small enough that all of those people are already "vibe coding" and, I mean, call me when any AI-generated game is actually worth playing.