How to Use Photoshop 27 (2026) With AI

Guide for Beginners/Professional

Adobe Photoshop CC AI helps you edit images faster without giving up control. You can remove unwanted objects, add new items, replace backgrounds, extend an image, improve image quality, and create complete visuals from a written prompt.

The best part is that you do not need to be a skilled designer to begin. If you can select part of an image and explain what you want in simple words, Photoshop can do much of the heavy work for you.

Still, AI does not replace good judgment. It may create strange hands, mismatched shadows, incorrect text, or objects that do not fit the scene. The strongest results usually come from combining AI generation with Photoshop’s normal selection, masking, color, and retouching tools.

What Is Adobe Photoshop AI?

Photoshop AI refers to the artificial intelligence tools built into Adobe Photoshop. Many of these features are powered by Adobe Firefly.

The main AI tools include:

• Generative Fill
• Generative Expand
• Generate Image
• Remove Tool
• Remove Background
• Harmonize
• Generative Upscale
• Generate Similar
• Reference Image

What Do You Need Before Using Photoshop AI?

Before you start, check the following:

• Install the latest version of Adobe Photoshop.
• Sign in with your Adobe account.
• Use an active internet connection.
• Check the number of generative credits available with your Adobe plan.
• Work with a clear, reasonably high-quality image.
• Save an original copy before making major changes.

Most generative AI actions use cloud processing. This means Generative Fill or Generative Expand may not work correctly without internet access.

How to Use Generative Fill in Photoshop

Generative Fill is one of the easiest ways to use AI in Photoshop. It can add, replace, or remove content from a selected area.

Step 1: Open your image

Start Photoshop and select File > Open. Choose the image you want to edit.

Before making changes, duplicate the original layer:

• Open the Layers panel.
• Right-click the image layer.
• Select Duplicate Layer.

This gives you a clean backup inside the same Photoshop file.

Step 2: Select the area you want to change

Use a suitable selection tool, such as:

• Selection Brush Tool
• Object Selection Tool
• Lasso Tool
• Marquee Tool
• Quick Selection Tool

Your selection tells Photoshop where it may generate new pixels. A clean selection often produces a cleaner result.

Step 3: Select Generative Fill

After making the selection, choose Generative Fill from the Contextual Task Bar. If that bar is missing, open it through:

• Window > Contextual Task Bar
• A prompt box will appear.

Step 4: Write a short prompt

Describe what you want Photoshop to place inside the selected area.

For example:

• Small wooden table with natural shadows
• You do not need to write a long story. Mention the object, material, color, angle, or lighting only when those details matter.
• Select Generate after entering the prompt.

Step 5: Review the results

Photoshop normally gives you several variations. Check each one and choose the result that fits the image best.

The generated content appears on a separate Generative Layer. Your original image remains available underneath it, which makes the edit easier to change or remove.

According to Adobe’s Generative Fill guide, the tool can alter selected areas through written prompts while keeping the generated edit separate from the original image.

How to Remove an Object With AI

You can remove a person, wire, sign, vehicle, stain, or other distraction in several ways.

Method 1: Use Generative Fill

1. Select the object and a small amount of the area around it.
2. Choose Generative Fill.
3. Leave the prompt box empty.
4. Select Generate.
5. Choose the cleanest variation.

Leaving the prompt empty asks Photoshop to study the nearby pixels and decide what should replace the selected object.

Method 2: Use the Remove Tool

• Select the Remove Tool from the toolbar and brush over the unwanted object. Photoshop will attempt to rebuild the area using information from the surrounding scene.
• For better results, make your brush slightly larger than the item you are removing. Cover the object fully, including its shadow if that shadow should also disappear.
• Small objects are usually easier to remove than large objects covering detailed backgrounds.

How to Add an Object to a Photo

Adding an object requires a clear selection and a specific prompt.

For example, suppose you want to place a lamp on an empty desk:

• Use the Lasso Tool to select an area on the desk.
• Select Generative Fill.
• Enter a prompt such as:

Small black desk lamp, side view, warm light

• Select Generate.
• Review the variations.
• Refine the best result with a layer mask if needed.

Make the selection close to the size and shape of the object you want. If you select half the image for a small lamp, Photoshop may generate an object that is too large or place it incorrectly.

How to Replace a Background With AI

Photoshop can separate the subject and create a new background around it.

1. Open your photo.
2. Choose Select Subject from the Contextual Task Bar.
3. Invert the selection by choosing Select > Inverse.
4. Select Generative Fill.
5. Describe the new background.
6. Select Generate.

A useful prompt might be:

• Bright modern office, soft window light, blurred background
• Keep the lighting direction in mind. A subject lit from the left may look unnatural against a background lit strongly from the right.
• After generating the background, check the edges around the person’s hair, clothes, and hands. Use Select and Mask when those edges need further cleaning.

How to Extend an Image With Generative Expand

Generative Expand lets you make an image wider or taller while Photoshop creates content for the new canvas area.

This is useful when:

• Turning a portrait image into a landscape banner
• Creating room for website text
• Changing an image for a social media size
• Fixing a crop that is too tight
• Adding more sky or background space

To use it:

1. Select the Crop Tool.
2. Drag the crop handles beyond the original image.
3. Choose Generative Expand from the Contextual Task Bar.
4. Enter a prompt or leave the prompt empty.
5. Select Generate.
6. Pick the best variation.

An empty prompt works well when you simply want Photoshop to continue the existing scene. A written prompt is better when you want a particular element in the added space.

Adobe’s current Generative Expand instructions also allow users to select an available AI model before generating the extended area.

How to Create a Complete AI Image in Photoshop

Photoshop can also create an image from text without requiring an existing photo.

1. Create a new document.
2. Select Generate Image from the Contextual Task Bar.
3. Write a description of the image.
4. Choose any available style or reference settings.
5. Generate the image.
6. Select a result and continue editing it with normal Photoshop tools.

A clear prompt could be:

Product photo of a white ceramic coffee cup on a light wooden table, morning window light, cream wall, realistic shadows

After creating the image, you can use Generative Fill to change smaller sections instead of regenerating everything.

How to Use a Reference Image

A written prompt does not always explain a certain layout, color style, product shape, or artistic look well enough. A reference image can give Photoshop clearer visual guidance.

Depending on the tool and Photoshop version, a reference can guide:

• Style
• Composition
• The appearance of an object
• Colors and visual mood
• Generated variations

For Generate Image, Adobe allows reference images to guide either the visual style or the composition. With Generative Fill, a reference can help Photoshop create a more specific object or appearance.

Only use images you own, have permission to use, or are legally allowed to use. A reference should guide the result, not be used to copy another creator’s protected work.

Final Thoughts

The easiest way to learn how to use Photoshop with AI is to begin with one small task. Remove an unwanted object, extend a background, or add a simple item to an empty space. Review the generated options and then clean the best one with masks and normal editing tools.

Use AI to save time, but keep your own judgment at the center of the work. A good final image should look intentional, believable, and suitable for the person who will see or use it.