Prompt engineering 101 for images: Crafting detailed visual descriptions

Images are a key part of creating engaging content for communicating your nonprofit’s mission. Social media, presentations, training documentation, diagrams to communicate ideas, email, and more – you can enhance your content with visual storytelling. With AI, you don’t need to be a designer to create compelling images: Copilot can be your collaborator in creating visual stories and elements.

Prompt ingredients for images are similar to how to prompt Copilot for content: Provide a description of what you’re trying to create “as if you were asking a helpful coworker”. Unlike content prompts, you can include lots of stylistic and visual details to build the image you need.

In this blog, we’ll explore how to hone our skills in describing images effectively. We’ll break down the essential ingredients for creating vivid visual descriptions to get the images you want.

Here is an image created using Microsoft Copilot Designer:

An AI-generated image of a dog with a green ball next to a tree created using Microsoft Copilot Designer.

Let’s explore the key components of the prompt used to create this image: 

Create an image of an excited brown terrier sitting near a tree with a ball. The ball is green and the ground is wet. Create the image in the style of a hyper-realistic drawing.

Prompt ingredients for images

1. Detailed subject
• The detailed subject is the central focus of the image. It’s what captures the viewer’s attention. 
• For example, consider an image of an excited brown terrier. The terrier’s expression and body language convey its excitement. 
2. Scene or setting
• Describe the environment in which the subject exists. What’s the backdrop? What’s the context? 
• In our example, the scene is outdoors. The terrier is near a tree, suggesting an open space. The ground is wet, indicating recent rain. 

3. Additional elements
• These are the details that enhance the overall image. They provide context and add depth. 
• In our case, the green ball is an additional element. The wet ground reinforces the recent rain. 

4. Specific style
• Consider the artistic style you want to convey. Is it realistic, impressionistic, or abstract? 
• Let’s choose a hyper-realistic drawing style for our terrier image. This means we’ll aim for intricate details. 

Crafting the image description

Remember, when describing images, engage the reader’s senses. Make them see, feel, and experience the scene. Whether it’s a playful terrier or a serene landscape, your words should paint a vivid mental picture.

Now, armed with these ingredients, let’s create stunning visual descriptions!

Keep learning

Explore these resources and learning paths to keep building prompt and image design skills. 

• Microsoft Digital Skills Center for Nonprofits: aka.ms/TechSoup
• AI skills for nonprofits collection on Microsoft Learn: aka.ms/AI-for-nonprofits-collection
• LinkedIn Learning pathways for digital and AI fluency: aka.ms/MyLearningPathway

Note: This article was created in collaboration between the author and Microsoft Copilot.

This is part two of a three-part series on AI prompts. Get notified when part three is published by hitting the Subscribe button with the in the upper right corner of the Nonprofit Community Blog space. 

Continue the conversation by joining us in the Nonprofit Community! Share your prompt examples and the images you created in the comments below.