The future of Construct

Oh, I see that any comment defending AI tools gets downvoted, so let me explain my point in more detail.

If you're developing a game for commercial purposes, you'll face tasks like:

1. Adapting your game for different platforms (web, mobile, etc.).

2. Long-term support of the game (fixing bugs, releasing quick patches, and so on).

3. Handing the project over to someone else who has no experience with Construct 3.

Many people are afraid that AI will take away the creative side of game development. But there's also a huge amount of routine work that has nothing to do with creativity and simply wastes developers' time.

If you're completely against using AI tools in Construct 3, chances are that game development is more of a hobby for you than a profession.

No one is arguing that AI doesn't do anything useful. The issue is the cost. What is the price we are paying for these conveniences? As a society, as working people, as creatives, and not to mention the environment.

I would guess that many downvotes come from people who recognize that AI is all but certain to have a net negative impact on humanity, regardless of what benefits it provides along the way.

It's also important to think of things on a larger scale, rather than an individual one. If every developer's productivity increases across the board from AI, then no actual benefit has been provided. Games in that case must simply be produced at a faster rate as the new baseline, which doesn't actually help anyone.

Increasing the rate of production also lessens the value of what is being produced, since the demand for those products does not increase alongside it. So professionals can have a negative view of AI just as much as a hobbyist, for this reason and many more.

For AI to be a positive advancement overall, I think many fundamental changes in human civilization will have to occur, and the problem is all the humans who will suffer along the way to make that happen, if it's even possible at all.

No one is arguing that AI doesn't do anything useful. The issue is the cost. What is the price we are paying for these conveniences? As a society, as working people, as creatives, and not to mention the environment.

I would guess that many downvotes come from people who recognize that AI is all but certain to have a net negative impact on humanity, regardless of what benefits it provides along the way.

It's also important to think of things on a larger scale, rather than an individual one. If every developer's productivity increases across the board from AI, then no actual benefit has been provided. Games in that case must simply be produced at a faster rate as the new baseline, which doesn't actually help anyone.

Increasing the rate of production also lessens the value of what is being produced, since the demand for those products does not increase alongside it. So professionals can have a negative view of AI just as much as a hobbyist, for this reason and many more.

For AI to be a positive advancement overall, I think many fundamental changes in human civilization will have to occur, and the problem is all the humans who will suffer along the way to make that happen, if it's even possible at all.

I think your view of AI is too disconnected from reality and from how AI tools are actually used in game development. You're thinking in pure abstractions — your opinion of AI is shaped more by social media than by real hands-on experience.

Sure there's uses to it theoretically, and it's not like I've never used it for some random small pieces of my code (it's great for writing regex for example because who wants to deal with regex...) and it's unironically awesome for that. But on the flipside we have seen:

1. Game developers showcasing their game for the first time and have 7 slop copies pop up on steam within 48 hours. Not that this wasn't a thing before, but with AI being as efficient as it is, sharting out a copycat project is so fast you can probably release it before the original whereas before it would at least require actual time and effort.

2. Game developers being blackmailed with threats like "I'll copy your game with AI unless you pay me 200$"

3. Game developers deciding against releasing a demo of their game in fear of it being copied with AI.

To top it off, clearly a large chunk of people is not using AI as a selective tool, and instead as a shortcut to circumvent any kind of effort, often in the pursuit of money. Paired with the speed that AI works at, I expect an absolute slop tsunami, and the only safeguard for it is that people oppose it on principle (and this will not work indefinitely). At the very least, the average quality of games will drop, at the cost of quality games possibly being drowned out and have copies of copies of copies being vomited out over night.

Don't look at what is theoretically possible with AI, look at what people actually do with AI, and it doesn't look good. I recently just got reminded of the whole rabbit hole of "AI generated drawing tutorials" (???) which is a whole thing. You cannot even google for arm anatomy without google proudly presenting you something like this as one of the first results.

AI is one thing on paper, but in reality it is nothing but an any% enshittification speedrun.

If you've made a game that's easy to copy with AI, then your game is worth close to nothing. I think we're past the time when games had value just because they exist. If you publish a game on Steam, Google Play, or any other platform, whether you get players or not will be decided by that platform's AI recommendation algorithms.

Like it or not, it's the players who decide which games they enjoy playing and which they don't. Bad games will have worse metrics (playtime, retention, and so on), which means the algorithms will recommend them less to new players. And nobody cares whether you used AI during development or not.

Meanwhile, AI lowers the barrier to entry for solo developers and small teams, speeds up prototyping, automates tedious tasks, helps with debugging, localization, testing, and lets developers experiment with ideas that would otherwise be too expensive or time-consuming to pursue.

This topic doesn't just have good posts — they're great posts!

It still takes lots of time on codex to make a game with openai. You have to iterate and play it and see what doesn't work and what's not fun. The thing you use the AI for most turns out to be playing your game, not writing it. Having someone (the ai) there to playtest for bugs or balance at 2 am is very useful. I think most of you should try to do a game with ai, so you will see it is still hard to come up with a good game, you just spend your time doing the hard part. You realize that the AI doesn't understand what fun is. It will dutifully play and write a dull game if you let it.

yours

winkr7