I’ve worked with a few folks who have been given AI tools (like a designer who never coded in his life, a or video/content creator) who have absolutely taken off with creating web apps and various little tools and process improvements for themselves thanks by just vibecoding what they wanted. The key with both these individuals is high agency, curiosity, and motivation. That was innate, the AI tooling just gave them the external means to realise what they wanted to do with more ease.

These kinds of folks are not the majority, and we’re still early into this technological revolution imo (models are improving on a regular basis).

In summary, we’ve given the masses to “intelligence” but creativity and motivation stay the same.

If you look at every game dev forum in existence, or you’ve ever talked to people about why they got into CS there are probably 1000x more people who want to publish a game than have done it.

If there was a a tool that provided a 10x-100x speed boost it would push enough of those people over the edge and make a significant impact on number of games released.

That’s to say nothing of boosting existing game devs.

Even if models stopped improving today, it'd take years before we see the full effects of people slowly gaining the skills needed to leverage them.

There doesn't need to be any "magic" there. Just clearly state your requirements. And start by asking the model to plan out the changes and write a markdown file with a plan first (I prefer this over e.g. Claude Code's plan mode, because I like to keep that artefact), including planning out tests.

If a colleague of yours not intimately familiar with the project could get the plan without needing to ask followup questions (but able to spend time digging through the code), you've done pretty well.

You can go over-board with agents to assist in reviewing the code, running tests etc. as well, but that's the second 90%. The first 90% is just to write a coherent request for a plan, read the plan, ask for revisions until it makes sense, and tell it to implement it.

But the big models have come a long way in this regard. Claude + Opus especially. You can build something with a super small prompt and keep hammering it with fix prompts until you get what you want. It's not efficient, but it's doable, and it's much better than having to write a full spec not half a year ago.

LOL: especially with Claude this was only in 1 out of 10 cases?

Claude output is usually (near) production ready on the first prompt if you precisely describe where you are, what you want and how you get it and what the result should be.

Nothing new here. Getting users to clearly state their requirements has always been like pulling teeth. Incomplete sentences and all.

If the people you are teaching are developers, they should know better. But I'm not all that surprised if many of them don't. People will be people.

Once people have had the experience of being a lead and having to pass tasks to other developers a few times, most seem to develop this skill at least to a basic level, but even then it's often informal and they don't get enough practice documenting the details in one go, say by improving a ticket.

Maybe it’s needing to step back and even ask for design doc before a plan, but even then…

But there are thousands of people on social media claiming huge productivity gains. Surely at least 5% of devs are holding it right.

If a 10x boost is possible, we’d notice that. There are only 20k games a year released on steam.

If my hypothesis is true and the real final output boost is somewhere near 20%, we’re seeing exactly what you’d expect.

There aren’t noticeably more total startups or projects though.

Eg, ai is a big multiplier but that doesnt mean it will translate to “more” in the way people think.

As an independent game developer myself, who has extensively tried to use LLMs in every which way for desire of a productivity boost, what they actually do is introduce code debt so quickly that the code collapses in on itself before you even get far enough done to ship an update, unless you handhold it so tightly that it would be faster to write the code yourself.

Now if it’s something closer to 20%, we’re seeing exactly what you’d expect.

But my guess would be: games are closed sourced and need physics. Which AI is bad at.

Many games don’t need physics, and there are a billion hobby projects on GitHub.

Does not look like less games.