The virtual world parts such as video and 3d, and plausible physics, I think we are going to do as well as image/audio/text within the next 10 years. Maybe even

The virtual world parts such as video and 3d, and plausible physics, I think we are going to do as well as image/audio/text within the next 10 years. Maybe even 5. These things primarily need to be believable and mostly-not-directly-wrong to serve a lot of usecases. And the way to get it to that level seems to be "just train it on Internet scale amounts of data". Whether we will really have cracked the physical world connections, of physics, genetics, etc that we can use it to make physical products, changes etc I am less sure. Many usecases like medicine require not just correctness, but also a degree of verifiability. It is being worked on a lot, with many promising results. But the just-scale-the-training data strategy seems less viable here, both because relevant data is less prevalent and may not give the level of correctness.