Generative Fill is a powerful AI tool built into Photoshop. We can use Generative Fill to remove objects in much the same way that the Remove tool works, but the real power of Generative Fill is that we can use it to create objects that have nothing to do with our current image. For example, I could use Generative Fill to add a unicorn to my Colosseum image...

 Or I could add a rocket ship...

Or an Egyptian pyramid...

Or even a giant gorilla riding a tricycle...

Notice that for each of the 4 images above, Generative Fill even created accurate shadows that matches the other shadows in the image

In fact, I can use Generative Fill to add anything I can think of to my image. Yes, this is fun to play with, but remember that we are interested in removing things from our image. Before we move on, notice that not only did Generative Fill accurately create what I wanted, but it specifically adjusted the colors, lighting, shadows, tint, and so forth, of the objects so that they look like they belong in the image.

So, how does it work? When using Generative Fill, we simply make a selection and type into the Generative Fill text box exactly what we want Photoshop to generate. The objects we request are then created using Adobe Firefly, which is Adobe's AI engine. The ability to create virtually anything we can image means that Generative Fill is an incredibly powerful Photoshop tool.

Ok, but can Generative Fill be used to simply remove objects? Turns out.....sort of. Keep in mind that the function of Generative Fill is to add or replace objects using simple text prompts to create realistic results. We can use it to add pretty much anything we want, and to also replace things that are in our image that we would prefer be something else (this is how we are going to use Generative Fill in this tutorial). If we simply use Generative Fill to remove something, Photoshop will actually apply the Remove Tool that we have already used.

We are going to use Generative Fill to its full potential by removing the round area below...

and replacing it with a nice, cool fountain.

The first thing we need to do is make a selection of the round area.

To use Generative Fill, we first need to locate the Contextual Task Bar. Look around your Photoshop window and locate the Contextual Task Bar...

If you do not see the Contextual Task Bar, activate it in the Window menu...

To generate our fountain, we simply need to click on the Generative Fill text area and type a description of what we want.

8. Click on the Generative Fill area...
   
   It should now look like this...
   

We now need a good description of what we want to see. Photoshop will then generate 3 different images that match our description. Keep in mind that the more specific our description, the closer we will get to what we want. For example, if I just type fountain as the prompt, this is what I get...

As you can see, these are pretty generic fountains. We need to use more descriptive terms. I'm going to use the term Roman fountain and see what I get.

9. In the Generative Fill text area, type Roman fountain...
   
10. Press Enter - Photoshop will think for a moment and then generate the fountains

You should immediately notice a few things. First, there is a fountain where our selection area was...

Note that your fountain will look different from mine. When Photoshop applies Generative Fill, it does so in a unique way for each user, so the fountains I get will be different from the fountains you get. Also note that the Properties panel has opened...

Pay attention to the Variations section at the bottom. Here you will see the three different fountains that Generative Fill has created. Clicking on each one will place it in your image.

11. Click each fountain variation so that it is placed in your image to see what each looks like

Second, you should notice that you have a new layer that contains your fountains...

Having Photoshop create a new layer means that our generated objects can be viewed or not, and are added non-destructively to our image. It also allows us to make changes to only the generated object if we wish.

Turns out that I'm not a big fan of any of the fountains Photoshop has to offer. There are two different ways we can get more generated images. First, we can click the Generate button on either the Contextual Task Bar (where we typed the text Roman fountain), or on the Properties panel.

12. Click the Generate button (in either place) to generate 3 new images - notice that the 3 new images are placed on the Properties panel...
    

All of the fountains that Generative Fill has created so far appear too modern for our image. I'm going to try a different description and see what happens.

13. In the Generative Fill text area, add the word Ancient to the description...
    
14. Press Enter to generate new fountains...
    

Looking through these new fountains, I like the second one in the image above, which looks like this in the image...

Keep in mind that your generated fountains will be different from mine. While I found one that I like (notice that the fountain in the image above looks very similar to the description of the actual fountain that was in place during Roman times), you may still have objects that you are unhappy with. This is not a problem, you simply need to continue generating fountains, or perhaps changing your description text, until you have a fountain you like.

15. Find a fountain that you like and select it to include it in your image

Let's save our work up to this point.

For our final individual tool, we will be using the Clone Stamp Tool, which is kind of a tool of last resort.