Generative AI in Photoshop 2024

Generative AI in Photoshop is powered by Adobe Firefly, a machine learning model that can generate realistic or surrealistic images based on your input. You can use it to explore and experiment with different ideas, create unique composite images, or make complex edits and refinements.

To use Generative AI in Photoshop, you need to have the latest version of Photoshop installed on your computer. You can download it from the Adobe website or update it from the Creative Cloud app. Once you have Photoshop ready, you can access Generative AI features from the Contextual Task Bar that appears when you select an object or area in your image. There are two main features: Generative Fill and Generative Expand.

Generative Fill allows you to fill a selected area with content that matches the rest of the image or with something completely different, depending on what you type in the text prompt box. For example, you can type “a cat” to generate a cat in the selection, or leave it blank to let Photoshop fill it with something similar to the surroundings. You can also use adjectives and adverbs to modify your prompt, such as “a cute cat” or “a cat wearing a hat”. You can generate multiple variations by clicking the Generate button again or by changing your prompt. The generated content is added on a new layer, so you can edit, refine, or remove it without affecting the original image.

Generative Expand allows you to extend the width or height of your image by generating content that blends seamlessly with the existing image. For example, you can turn a vertical shot into a horizontal one, or add more sky or ground to your image. You can also use text prompts to specify what you want to see in the expanded area, such as “a mountain” or “a city skyline”. You can adjust the aspect ratio and the direction of the expansion by dragging the handles on the image. The expanded content is also added on a new layer, so you can edit, refine, or remove it as you wish.

Generative AI in Photoshop can help you create amazing images in a fraction of the time it would take if you did it manually. However, there are some tips and tricks that can help you get the best results. Here are some of them:

  • It’s OK to leave the prompt blank. Sometimes, you don’t need to specify what you want to generate, and Photoshop will do a good job of filling or expanding the image based on the context. This can be useful when you want to remove unwanted elements or extend the image without changing the scene.
  • Be creative with your selection. You don’t have to select a rectangular or circular area to use Generative AI. You can use any selection tool, such as the Lasso or the Pen tool, to create a custom shape for your selection. This can give you more flexibility and control over the generated content.
  • Don’t use instructional prompts. Generative AI works best when you use descriptive words, not commands or questions. For example, instead of typing “remove the car”, type “grass” or “road” to generate something that replaces the car. Similarly, instead of typing “what if there was a moon”, type “a moon” to generate a moon in the image.
  • Utilize the Object Selection tool. The Object Selection tool is a handy tool that can automatically select an object in your image with a single click or drag. You can use it to quickly select an object that you want to fill, expand, or remove with Generative AI. You can also refine the selection by adding or subtracting areas with the other selection tools.
  • Use it to blend photos together. Generative AI can also help you create stunning composite images by blending two or more photos together. For example, you can select a person from one photo and paste it onto another photo, then use Generative Fill to generate a background that matches the person. Or, you can select a part of an image and use Generative Expand to generate a larger image that includes the selected part.