The latest algorithm for Generative Fill seems suddenly dumber?

I strongly agree. I have the same problem—after inserting an image without a caption (due to the new update that removes captions), the results become really poor.

YES! What is up with this?! It's gotten insane! Need to fill in a shadow? Here's a horror face or squirrel, or non-sensical tropical bird! I almost wonder if competitors are messing with the algorithm or something?! It used to fill patterns beautifully! WHAT THE HELL.

I have updated to photoshop 26.0.0 and the results of generative fill are absolutely beyond usable.
I was able to reinstall 25.12.0 and It works just fine.
The quality of the new release 26.0.0  has dropped to an unusable level. Textures that were rendered beautifully in previous versions now sometimes appear smudged or as if painted in with a brush. There are far more nonsensical fills—like birds randomly appearing in hair or clothes, and extra hands in bizarre places when the image already has two! It’s like AI gone wrong. What happened to quality control? How does the AI actually get worse? This is beyond frustrating.
To make things worse, the newest update deleted my older versions, even though the "delete older versions" checkbox was unchecked. It's incredibly frustrating to lose those versions without consent.

I would echo what Pete said about leaning into the use of a reference image (works best with a clear subject on a white or black background). By providing a bit of additional input into the model you should get much better results. 

This verison of the model was trained to give more variability in the output but higher prompt adherence (you should be getting cowboy hats but more variety in them). Clearly we have further to go in getting to that objective but use of the reference image should get you back on track. 

Btw - if you are interested in sending me the files (PSD including the masks and prompts and generative layers created) that are proving to generate problematic results. I'm happy to bring them to the attention of the people building out the latest model so they can investigate why the results are poor. 

Hi @Jen27654846oo5f ,

Are you using the beta or the release Photoshop?

With beta, you can also use Reference Image to point to the type of hat you're after as well. 

If either one, you can also use the thumbs up and down to report generations as positive or negative in quality or in result. That's great feedback for the team as they work to improve these generative models.

Regards,

Pete