I'm curious about the timeline for AI to be able to create clones of older games, like Super Mario World, but with a fresh theme and maybe some hidden levels. We all acknowledge it might take a while before we can request something complex like a hack-and-slash MMO in a Diablo-style world. But I'm wondering how long it might really take for AI to effectively handle simpler, vintage games?
8 Answers
There are still issues with consistency and customization, but since AI could train on existing games, I envision something by around 2027 for decent results.
Using tools like Claude or ChatGPT for simple games is super easy. Everything is advancing so quickly that I wouldn't be shocked if we get a suite of tools for game creation by year's end and something really robust by 2027.
What if we reached a point where games could be generated on the fly based on your input? That would be revolutionary!
I’m curious about what these games will actually look like. Are we talking more like Pac-Man or something in-depth like GoldenEye?
Honestly, I think if you put in enough research and effort, you could get it to work as soon as next year. By 2030, I wouldn't be surprised if you could generate full AAA games just by giving an AI a good prompt.
It's mind-blowing to think about! I wonder how this will impact the gaming industry—AI might significantly change how we develop and test games.
Look at the Great Giana Sisters—it's essentially a clone with its own spin. It shows it's already possible to create engaging games using AI.
The timeline for AI advancements is always 2-5 years for anything, and I believe that’s just how it will be moving forward.
Actually, some AI can do this now! Over the past year, I've been using various AIs to create JavaScript/HTML/CSS versions of old games, like a version of Missile Command. It's incredible what can already be done.
Those might be simpler than Super Mario World, though. The complexity difference is huge.
I think we could see basic AI game generation within a year. For something impressive, maybe around three years, and possibly five years before it truly shines.
I agree! If you consider the pace of AI development, it may not take as long as we think.
I’d say halve those estimates. One year for basic stuff sounds about right, but three years? With rapid advancements, I’m skeptical about that.
If the prompt is detailed enough, you could potentially get AI to create something right now!
Copyright restrictions may limit AI, so we might only get games 'in the style of' existing ones unless things change in the industry.