The tools on this site all make use of a large neural network called Stable Diffusion and capable of generating images from text. Stable Diffusion is one of the largest Open Source projects in recent years, and the neural network capable of generating images is "only" 4 or 5 gb heavy.
Since the neural network is nothing more than a mathematical model that most likely completes all the pixels in the image, it is also possible to make editing changes by giving the image to the neural network and making the change upstream via text so that the image is in fact edited.
It is possible, as in the case of inpaint, to give the neural network an image where a part has been deleted, and let it complete it according to an instruction we have given it.
Similarly, it is possible to generate infinite "variations" of any image, since the neural network is able to tell us what to it is most similar to our image.
As good as artificial intelligence is at drawing, it is not conscious in the current sense of the word. Intelligence is a purely mathematical fact, that is, it measures the ability of the neural network in being able to perform complex tasks, such as it may be able to talk to you or generate images.
However, they are only complex mathematical algorithms, nothing more. Consciousness is something typical of living beings, not machines.
.jpg)
 Cropped.png)
.jpg)
.jpg)
