Building an AI game studio: what we've learned so far

Something that I never see addressed for this: So, let's just grant that someday, the tech will be mature enough that this is possible, and let's even say it goes beyond videogames to movies, to visual art, to graphic design, to writing, etc. Let's say that AI gets to a place where any joe blow can put in a prompt, and get a competent, and even let's be generous and say good product out of it. A solid 8/10.

So... who the hell is going to buy it? Because videogames as an industry is already entirely saturated with products that range a whole spectrum from utter dogshit to amazing works of technical expertise, writing, design, etc. There are over 70,000 games on Steam alone now, with 9,000 added in the last 9 months. If this tech actually got to this place, there will be exponentially more games, because all you have to do is tell an AI what you want to play.

And you can take that further: Movies are also highly saturated as an industry, especially as larger studios move ever further into less making "movies" or "series" and just making "content" endlessly remixing their intellectual properties. So now, all of those companies (and all the people who like their stuff) can now just make their own Iron Man movie? Their own Wandavision? Just endlessly making and remaking and remaking, as though tons of people aren't already sick to death of all the television programs and movies that are being made?

And again, you can just keep extending this to any media: print, music, art... we have more of everything now than we ever have before and the goal of companies like Adobe, like OpenAI, etc. is to put even more powerful creative tools into even more hands, broadening the group of people who can create stuff but like... even if you take it as granted that this can be done...

Who the hell is watching all of this stuff? Who is playing all of these games? And why in the world would you pay to watch someone else's AI movie when you can pay to generate your own with whatever you want in it? Why would you ever buy a game off Steam again if you can just ask your game making AI to make you the exact game you want, even just copying the damn description out of steam?

All I see this doing is potentially killing off dozens of creative industries and funneling shit tons of creative control and platform-style power to a handful of massive corporations, running warehouses full of fucking graphics cards, to generate the same games, the same movies, the same music, over, and over, and over, to suit everyone's personal taste, and absolutely destroying entire rainforest's worth of electricity to accomplish it. And like... why do we want that?

> you will never be able to fix it.

I don't know about the article's studio specifically, but regarding AI game engines, I'm mostly worried you won't be able to make engine edits, or the engine will only be able to make an extremely narrow subset of games, or exiting the WYSIWYG/SaaS means starting over from scratch. If the engine targets a specific genre, it could cause a flood of low effort garbage. Some genres are already flooded because the easier game engines have tutorials and asset packs for them.

If AI works well enough that just a vague prompt leads it to spit out a professional, compelling and creative game with assets, VO, music, coherent level and production design, and everything else that goes into successful modern games... companies are going to own that and keep that on lockdown, because it's essentially a free money machine.

You won't have a "game making AI" that isn't already owned by big media companies, crippled and handicapped, and expensive as hell. You won't legally be allowed to compete against them. ChatGPT is never going to do that. That isn't how capitalism works.