OpenAI Working on Social Network With Image Generation Features

[MacBH928]

flickr 2.0

 

[Oestreich Dieter]

You people are missing the point, its primary purpose will be for training the LLM on the data generated by the social network. They arent actually trying to compete with FB or IG.

 

[1037]

I'll stick with X

 

[Jean Claude]

Perhaps you are over thinking this.

You as an individual human looks at an image.
Based upon your long established preferences in life, what you like and enjoy, you view the image and feel indifferent, or you dislike it, or you like it.

Why does it need to be any more complex than that?
No further thinking should be required.

This is the problem with most of the art world.
It's not what it is, it's who did it that we use to add or reduce value.

Show an art expert a picture and if they find out it was painted be a young child just messing around, it's rubbish, but if it's painted by some long dead person who was famous for various things in his life, it's amazing.

Eh, pretty much missing the entire point of my comment. It's not "who/what" is creating it (I literally said in my original comment I was quite impressed with the art). It's the proliferation and commonality of it all that has cheapened it to parody status. If there were millions of Renaissance Masters creating myriads of masterpieces across Europe, the same exact thing would have happened.

 

[jedivulcan]

I like using AI tools to generate hash tags for things I’ve written in a social media post

or simplify what I’m saying so it’s easier to read for a social media post.

I think whatever they’re experimenting with in Sora still needs a lot of work from a moderation perspective but the remix feature is a lot of fun.

 

[Graphate]

I’m not sure it’s a good idea to get involved with what many people believe to be the worst future thing to happen to humanity, involved with what many people believe is currently the worst thing to happen to humanity.

Or maybe tying the two together will end up going great.