To answer another question about backflip.ai, I personally have not used it myself. I do work in software, and use AI on a day to day basis. The fact that they use a foundation model gives me hope that it is good at what it claims. Probably worth throwing $20 at it with a few projects in mind. I would think converting 3d scans to CAD is the ultimate use case for this.
I agree, AI seems to be taking these kinds of applications in a positive direction. My guess is not slam dunk but even if it does some of the heavy lifting processing in a reasonable manner, might be worth it. Fom what little I've lurked around 3D scan projects, it seems to be one of those -great now I have a point cloud, now the real work starts. If its 'just' a shape like a garden gnome, nobody will see the differnce. If its a mechanical thing with specific dimensions, the operator has to oversee & edit those issues. So if auto pilot get 90% correct & designer is left to alter 10% thats a win. But if designer ends up doing 70% of the work to fix or fudge autopilot output, the needle swings the other way.
Loose comparison. I've been checking out different photo apps. Its another industry with AI being name dropped like crazy. In some cases its just gimicky, but in other cases like so called simple masking, its amazing what AI figured out compared to the hours spent by a human. A couple of brush strokes to tell AI this is the subject or this is the background & away it goes. Zoom in at the pixel level & its masking incredibly fine detail. And then it just goes from there. And with that comes the processing specs. Processor/ram/GC... need to be quite robust to handle the crunching.

