AI Essentials Course — Phase 3: Intermediate Techniques
Session 15: AI Image Generation — Creating Visuals for Presentations
Use Google Gemini's image generation to create professional visuals for academic presentations — and learn the art of writing precise, descriptive image prompts.
Learning Objectives
What You'll Learn
By the end of this session, you will be able to:
- Use Google Gemini to generate images from text descriptions
- Write effective image prompts that specify style, content, mood, and composition
- Iteratively refine an image prompt based on the initial result
- Understand the ethical considerations of using AI-generated images in academic work
Platform Access
Getting Started with Google Gemini
Follow these steps to access Google Gemini and get ready for today's lesson.
- Go to https://gemini.google.com and sign in to your Google account.
- Start a new conversation.
- Image generation in Gemini is triggered by asking it to create, draw, or generate an image in your prompt. You don't need to click a special button — simply describe what you want.
- Note: If Gemini cannot generate an image for your request (sometimes content filters apply), simply rephrase your description to be more general or academic in nature.
- Think about a hypothetical academic presentation you might give — perhaps introducing your research area, explaining a concept, or illustrating a process. You'll create images for that presentation today.
- You're ready to begin. Image generation is one of the most visually engaging sessions in the course — enjoy the creative process!
Free Account Required
All platforms used in this course offer free accounts with no credit card required. If you already have an account, simply sign in. The free tier gives you everything you need to complete this session.
Core Lesson
Today's Lesson
Read through this lesson carefully before starting the practice exercises below.
A picture is worth a thousand words — and in academic presentations, the right image can make the difference between an audience that's engaged and one that's struggling to follow your argument. Today you'll learn to use AI image generation to create custom visuals that serve your specific academic needs, rather than relying on generic stock photos or clip art that may not match your content.
AI image generation works by taking a text description — called an image prompt — and generating a completely original image based on that description. The more specific and detailed your description, the more accurately the AI can produce what you envision. This is a skill that takes practice, and today's session is designed to help you develop it through experimentation and refinement.
Effective image prompts typically include several key elements: the subject (what you want to show), the style (photograph, illustration, infographic, diagram, watercolor, etc.), the mood or atmosphere (warm and welcoming, professional and clean, dramatic and bold), and the composition (close-up, wide-angle, centered, landscape orientation). Adding these elements transforms a vague request like 'an image about learning' into a precise description that produces something genuinely useful.
For academic presentations, AI-generated images are particularly valuable for three types of visuals. First, conceptual illustrations: abstract ideas made visual — showing the concept of 'connection' as a network of glowing lines, or 'growth' as a plant emerging from books. Second, demographic representations: depicting the specific populations relevant to your research, such as older adult learners in a classroom setting. Third, process diagrams: showing a sequence of steps or a cyclical process in a clean, illustrative style.
Using AI-generated images in academic work comes with important ethical responsibilities. You should acknowledge that images were AI-generated, either in your slide notes or in a brief attribution statement. Images of real people should be approached carefully — AI can generate images that resemble actual individuals, which can be misleading. And many academic journals and conference programs have specific policies on AI-generated imagery — always check before submitting visual work for publication.
One of the most valuable things you can learn today is iterative refinement: generating an image, evaluating what's right and wrong about it, and then adjusting your description to get closer to what you want. This back-and-forth process of prompt, evaluate, and refine is the same iterative approach that works for text prompts — and it applies beautifully to image generation as well.
Hands-On Practice
Practice Exercise
Follow these steps in Google Gemini. Take your time — there's no rush. Learning happens through doing.
- Open Gemini and type: "Please generate an image showing an older adult student sitting at a desk with a laptop, looking engaged and confident. The style should be warm, professional, and photorealistic. Natural light coming from a window."
- Review the generated image. What's right about it? What would you change? Type a follow-up that refines your description based on what you observed.
- Try a conceptual image: "Generate an illustration representing the concept of 'lifelong learning' — show this as a symbolic visual with books, light, and a sense of growth and possibility. Style: modern flat illustration with warm colors."
- Try a process image: "Create an infographic-style diagram showing 5 stages of the dissertation writing process, from topic selection to defense. Use simple icons and clean typography. Blue and gold color palette."
- Review all three images. Which technique produced the most useful academic visual? What elements in your description made the biggest difference?
- Bonus: Try to generate an image that represents something specific to your academic field — a concept, a setting, or a population you study. Refine it through at least two iterations.
Try These
Example Prompts to Try
Copy any of these prompts directly into Google Gemini and see what happens. Feel free to modify them to match your own academic interests.
Summary
Key Takeaways
- AI image generation creates original visuals from text descriptions — the key is writing prompts that specify subject, style, mood, and composition.
- Iterative refinement — generate, evaluate, adjust the description, regenerate — is the most effective approach to getting the image you envision.
- AI-generated images for academic presentations should be acknowledged as AI-generated; check your institution or conference policies before using them in submitted work.
- For academic presentations, the most useful AI-generated image types are conceptual illustrations, demographic representations, and process diagrams.
Descriptive Visual Prompts — Specifying Style, Content, Mood, and Composition
You've learned to write image prompts that go beyond vague requests — specifying the subject, visual style, mood, and composition details that produce genuinely useful academic visuals. This descriptive precision skill applies to all AI image generation tools and will serve you in presentations, posters, and any visual academic communication.