Still always check the sources. One of the most useful thinks to know. Is the source trustworthy or not, where does the image come from…
This advice will be useless soon.
Point 2 is already irrelevant. The announcement today from OpenAI about DALL-E stated it can generate text as part of the query cleanly now. Let me see if I can find the link.
I’ll believe it when it’s actually released. The press release images probably cherry picked the best images out of who know how many iterations.
Yes, it is definitely giving me the uncanny vibes: looking at his white cap, seemingly made of the same plastic as that funny coat.
But all in all, the article is quite meaningless. Can’t do much with these ‘tips’.
The links in the article were interesting, the actual tips were pretty average
Also the shadows don’t make sense. The lighting is shining directly on the subject as though it was taken in a studio, but the photo appears to be taken outdoors
This was also generated nearly six months ago (more?). Between SDXL, the newer midjourney versions and DallE 3, this articles shelf life is practically nonexistent
A lot of it was just common sense. The irony is that article maybe AI generated itself.
Disappointed article did not provide a high res, uncropped version of the Pope in a puffy coat.
First thing you look at: the hands.
(photo does not show hands)
The generative ai have been enhanced with more training on the hands. They hot much better.
They did link to it though.
uhh… lets put it this way. it took humans how long to generate generative ai? how long do you think it will take to perfect it? im guessing a lot less time.
