Agree. Indeed it's still looking for the breakthrough. The question is whether it's just around the corner like a surprising threshold to usefulness (E.g. like

Agree. Indeed it's still looking for the breakthrough. The question is whether it's just around the corner like a surprising threshold to usefulness (E.g. like GPT models) or if it's still decades out. I think there aren't enough people who will want to wear it for games or social media. Not because it's a bad experience, but because it's an expensive and cumbersome experience and it's just not THAT much better than just staring at a small screen. The user base can't motivate spending billions on creating the experiences that would grow the user base significantly. What I think could be the breakthrough thing is the combination of AI and VR. Interactive agents and possibly also something to cheaply create that holodeck experience for rendering (NeRFs with foveal rendering, whatever - I'm an amateur skimmer of papers and extrapolating research is difficult even for experts). But imagine the difference in development costs between a VR experience where some interior space is modeled by hand in minute detail in Unreal engine, and NPCs are scripted, versus one where a space is merely captured in RFs and the NPCs are GPT bots with some plot instructions. Both of those technologies (NeRFs and GPT NPCs) would have sounded like science fiction even to the creators of the Quest2, yet here we are. I would have said "no one would ever buy/wear this" 5 years ago, but with the idea of a mind blowing photorealistic interactive murder mystery developed on a shoestring budget (Not to mention: porn - never underestimate it) I'm not so sure any more.