AI-generated 3D objects, which industry's gonna adopt it first, e-commerce or gaming?

The Forge / 工坊 Models / LoRA / 模型与 LoRA
1

For AI-generated 3D, I’m betting e-commerce will go mass-scale before gaming. E-commerce has a high tolerance for 3D quality—like a product preview or a rotatable display model, a few tens of thousands of polygons and passable textures are fine, since users just spin it around a couple times.

Gaming is way more demanding: clean topology, proper UVs, controlled polygon count, and it has to be rigged for animation. AI-generated models basically need a full rebuild before they can go into an engine.

So in the short term, e-commerce stuff like product modeling and AR try-ons will reap the benefits first. Game artists will mostly use AI for concepts and references; actually integrating it into the pipeline is still a long way off.

2

E-commerce is the first to cash in on this bet, but the tolerance for precision is way too different.

3

For games, the topology needs to be clean and UVs need to be neat. Stuff directly generated by AI will have to be scrapped and redone if you try to bring it into the engine.

4

Yeah, AR virtual try-on is pretty much ready to roll out. There’s a huge demand for product modeling.

5

Yeah, rigging bones for animation is something AI still can’t handle right now.

6

I make games, and AI models are at most just for reference. Redoing everything is the norm.

7

For e-commerce, as long as it looks decent at a few thousand polygons, it’s good enough. But for games, this quality is nowhere near the standard.

8

Man, I can’t keep up with all this new stuff anymore.