https://github.com/lucidrains/dalle-pytorch
It just needs to be trained
https://m.youtube.com/watch?v=9CO6M2HsoIA
Compare 2:33 in the above fictional video to this real one:
https://m.youtube.com/watch?v=DjUdVxJH6yI
https://www.thedrive.com/the-war-zone/35726/the-army-has-unv...
https://www.msn.com/en-us/news/world/attack-drones-dominatin...
These are previews of future wars, much like the wars in the decades before WW1 previewed the utter slaughter that was to come thanks to the new technological era of mechanized warfare.
Much more accessible to regular people.
Ability to be anonymous and hide in a crowd of drones.
Really? Their name says OpenAI.
Anyway, how do we know their examples were not cherry-picked? Do they have an online demo?
They have everything to lose by lying. If they say that these examples are not cherry-picked, then we have no reason (a priori) to doubt them.
On a side-note, the fact that you could doubt the results are real is telling: each of their compute-heavy experiments shakes our belief and further reinforces their reputation.
The singularity approaches.