Are AI Devs Thinking Things Through?
retained text
I’m not anti-ai, and I used to use it a lot for making prototypes to test ideas out, but the thing with programming with AI is that it can and will mess up the codebase. It’s a lot better than it was, I’ll give you that, but it still has pretty major flaws, plus the economic and environment impact but that’s not what I’m posting about. What I’m curious to see shake out is how AI developers will handle bugs post release, especially those that simply don’t know how to code, or what to look for, and are winging it. We all know that part of fixing bugs is creating more bugs, especially if your systems aren’t built around solid architecture. Even then sometimes bad things can happen. AI can only go off of what it is trained on, and the context you give it. So if folks using AI to ship games don’t know how to implement good architecture for their specific game, what are they going to do post release when bug fixing creates more bugs and the AI can’t find them? That’s so much wasted time, plus they’ll be paying money to use the AI to fix those bugs. Again, I’m not anti-ai but **some** of these developers are going to make a really, really bad name for themselves and have a bad time of things once their games are out for a length of time. Pushing patches out to real people is way different than play-testing a fix rigorously yourself or with a team.
document fetched 2026-11-16 00:24 · canonical https://reddit.com/r/gamedev/comments/1xx9uv3/are_ai_devs_thinking_things_through