[ad_1]
At this point, things get very technical (here you will find an explanation detailed but understandable). In a nutshell, the process consists of three steps: first of all, a text command is sent to a text encoder (text encoder) that has been trained to decipher it and assign numerical values to it. The second step instead involves the intervention of a model called “prior”, which associates the text encoding with a corresponding image encoding, acquiring the semantic information of the text command. Finally, an “image decoder” generates an image that is one visual manifestation of semantic information.
While the results are often amazing, this tool still has several critical aspects: “First of all, datasets can contain biases within them and this is why many companies are reluctant to freely market their tools”Perazzi always explains. “Imagine for example the misinformation that can potentially cause a system capable, by asking him, of inventing images of someone who is in a place where he shouldn’t be “.
Problems can also be of another kind. For example, it was noted that by asking to produce images of a nurse or assistant, Dall-E 2 created spictures of women onlywhile asking to create a lawyer or CEO they came out invariably of men. In short, our social prejudices end up in the images we produce and which then become the datasets used by the algorithms, which inevitably absorb and they in turn reproduce those same prejudices.
There are other aspects to be addressed: for example the risk that these tools overwhelm the creative industry, allowing anyone to create images without the need to involve any artist (think for example of record covers or advertising posters). For sure, there will still be a need for someone to think about what are the best commands to give to the artificial intelligence system e then select the resultsa process that – provocatively – could be considered not too different from the way in which Andy Warhol worked with his assistants.
“Of course, it is the human being who instructs the models“concludes Perazzi. “But a little information is enough, without the need to give details. I believe that tools like Dall-E 2 are partly tools and partly also authors, that these models have some kind of imagination. Sure, it can’t create something absolutely new, but we humans can’t do it either. In my opinion, there is also a creative element in the art of artificial intelligence“.
.
[ad_2]
Source link
