Their example videos: https://openai.com/sora/, of the doors opening, are hilarious.
1. The first set of doors doesn't have any doorknobs or handles. https://ibb.co/PwqfzBq
2. The second set of doors has handles, and some very large/random hinges on the left door. https://ibb.co/JkDtc6r
3. The third set doesn't have any handles, but I can forgive that, because we're in a spaceship now. The problem is that the inside of the doors seem to have windows, but the outside of the doors, doesn't have any windows. https://ibb.co/nwpXmtq & https://ibb.co/wr6v2g1
4. The best/most hilarious part for me. The doors have handles, but they are on the hinge side of the door. No idea how this would work. https://ibb.co/gWXDcfr
I think they trained on one too many closet bifold doors [1].
If you look at the edge of the doors as they swing open, it seems their movement resembles bifold door movement (there's a wiggle to it common to bifold doors that normal doors never have). Plus they seem to magically reveal an inner fold that wasn't there before.
[1]: https://duckduckgo.com/?t=h_&q=interior+bifold+closet+doors&...
There are more examples of its limitations.
The video with dogs shows three taxis transforming into one, the number of people under the tree changing https://player.vimeo.com/video/1037090356?h=07432076b5&loop=...
An example from the HunyuanVideo is terrible as well. Look at that awful tongue: https://hunyuanvideoai.com/part-1-3.mp4
And what we see in that marketing is probably the best they could generate. And I suppose it took a lot of prompt tweaking and regenerations.
The internet is already full of junk shorts and useless videos and soon there will be even more junk content everywhere. :(