Right? Came to the comments specifically for this, but am confused by people's responses. With prompt adherence this bad, is it worth the 2 cents you spent on it? I don't see how it's even useful for deciding if you want to use the ultra version, or for anything else really.... Maybe if you want to redo it in Photoshop? But at that point, breaking out the old Wacom tablet and making a composite image would probably be just as time intensive, but with much higher image quality (and none of the tale tell signs of AIgen)

Even if you only earn $12/hour, 2 cents is worth it to save just 6 seconds.

An image has to be much worse than that to fail to save you 6 seconds.

That said, this is their own chosen example of what it can do, so I'd have to assume it is much worse than that on average.

Will this save me 6 seconds? It'll take me longer than that to come up with a prompt, type it, enter it into the service, wait for it to generate, download it...

And again, if I can't use it because it's totally wrong, then... what are we even doing here?

> Will this save me 6 seconds? It'll take me longer than that to come up with a prompt, type it, enter it into the service, wait for it to generate, download it...

It will probably save a lot more, but the point is 6 seconds is the threshold at which 2 cents is "worth it".

Good art takes a long time to create.

If this image were representative, errors and all, it would be where you could expect a professional to reach after an hour or so, give or take — I've seen professionals working on an icon set for multiple days, and most webcomics I see, even when it's their full time job and they've got a good system going to make their output easy for themselves, don't tend to do produce outputs like this should have been more than once per day.

> And again, if I can't use it because it's totally wrong, then... what are we even doing here?

On this, I tend to agree. If you have a specific output in mind, quite often they're just wildly wrong. Repeated generations are just plain bad, and the system just can't seem to get what's being asked for.

> Imagen 4 Ultra: When your creative vision demands the highest level of detail and strict adherence to your prompts, Imagen 4 Ultra delivers highly-aligned results.

It seems that you may need the "Ultra" version if you want strict prompt adherence.

It's an interesting strategy. Personally, I notice that most of the times I actually don't need strict prompt adherence for image generation. If it looks nice, I'll accept it. If it doesn't, I'll click generate again. For creativity task, following the prompt too strictly might not be the outcome the users want.

I've found this is an interesting balance with Copilot specifically. Like, on the one hand I'm glad it aims for the bare minimum and doesn't try to refactor my whole codebase on every shot... at the same time, there's certain obvious things where I wish it was able to think a bit bigger picture, or even engage me interactively, like "hey, I can do a self-contained implementation here, but it's a bit gross; it looks like adding dependency X to the project keeps this a one liner— which way should it go?"

Give me a 'precision' slider then. On one end it should do precisely what you asked, to a T, even if what you asked for is dumb, and on the other end it should try to capture the spirit of what you wanted plus any obvious oversights.

I’ve had good experience with iterative prompting when generating images with Gemini (idk which model — it’s whatever we get with our enterprise subscription at work, presumably the latest.) It’s noticeably better than ChatGPT at incorporating its previous image attempt into my instructions to generate the next iteration.

Though that was only Imagen 4 Fast, not Imagen 4 or Imagen 4 Ultra.

Same for the poster. Asks for the ship to be going towards the right, and it's clearly doing the opposite

As seen from the AI's perspective.

To the left of the "detailed spaceship" I think I see a distortion pattern reminiscent of a cloaked Klingon bird of prey moving to the right. Or I'm just hallucinating patterns in nebular noise.

The ship is reminiscent of Galactica's oldschool vipers. Different, but very similar overall structure.

Hopefully it's better than midjourney at least. Ignoring key parts of the prompt seems to be a feature.

Midjourney scores the absolute lowest in terms of prompt adherence against any of the other SOTA models (Kontext, Imagen, gpt-image-1, etc). At this point, its biggest feature is probably as an "exploratory tool" for visualizations by cranking up the chaos and weirdness parameters.

In the little experimentation I did with AI image generation, it seems more a game of trying multiple times until you get something that actually looks right, so I wonder how many attempts they did.