How to create images with DALL-E or similar AI tools
Learn how to create amazing images with DALL-E and Gemini in minutes. Step-by-step guide for beginners with examples and effective prompt tips.
Transparency: This article contains affiliate links. If you purchase through these links, we receive a small commission at no extra cost to you. You support "La Bussola dell'IA," keeping it independent and free. Thank you!
Transform Your Ideas into Extraordinary Images Without Knowing How to Draw
Have you ever wanted to create a unique illustration for your blog, a social media post, or a presentation, but stopped because you can't draw? Or do you need original images for a project but don't want to spend hours searching for the perfect one among thousands of stock photos? Today, artificial intelligence can transform simple textual descriptions into stunning images, allowing you to visualize any idea you can imagine in seconds. In this tutorial, we will show you how to use DALL-E and similar tools like Gemini (Nano Banana) to bring your ideas to life, even if you have no artistic experience. A recent study by the Harvard Business Review" with a reference to the actual article "How Generative AI Can Augment Human Creativity" published in July 2023. This article indeed discusses how generative AI can enhance human creativity.
What AI Image Generators Are and How They Work
AI-based image generators are advanced systems that transform textual descriptions (called "prompts") into images. These models have "seen" millions of images during their training, along with their corresponding descriptions, learning to connect words and visual concepts. When you input a prompt like "a cat wearing a chef's hat cooking pasta in an Italian kitchen at sunset," the AI understands the individual elements (cat, chef's hat, kitchen, pasta, sunset) and combines them into a coherent image.
The two most popular tools in this field are OpenAI's DALL-E and Google's Gemini (with its image generation feature previously called "Nano Banana"). Both allow you to create unique and personalized images, but with some differences in the interface and specific capabilities. In this tutorial, we will see how to use both, starting with Gemini since we will use it for our tests. If you are interested in understanding the differences between these tools, webdomus.net has published a comparative analysis of the main AI image generators that highlights the strengths of each.
To work best with these tools, a monitor that reproduces colors faithfully is essential. Working on screens with inaccurate colors means creating images that will appear different to others. A quality monitor or a monitor light bar like the Quntis Light Bar can make a difference, eliminating glare and ensuring a clear and precise view of the colors you are generating.
INITIAL SETUP: Accessing Gemini
Gemini is Google's artificial intelligence model that includes image generation capabilities. Here's how to get started:
THE STEP-BY-STEP GUIDE
Step 1: Create a Google account If you don't already have a Google account, you will need to create one. Visit accounts.google.com and follow the instructions to sign up. If you already have an account, you can proceed directly to the next step.
Step 2: Log in to Gemini Go to the official Gemini website at gemini.google.com and log in with your Google account. Once logged in, you will be directed to the main Gemini interface.
Step 3: Write your image generation prompt In the text box at the bottom, write a request that begins with words like "generate," "create," or "draw" followed by a description of the image you want. Be as detailed as possible, specifying the subject, setting, artistic style, lighting, and other relevant elements.
Example of an effective prompt: "Create an image of a futuristic library with floating shelves, soft blue lighting, and librarian robots assisting visitors. Photorealistic style with dramatic lighting."
Experts from copywriterai.it advise always starting with a simple formula like
Organization is half the battle. To keep track of your most effective prompts and successful style combinations without cluttering your browser with tabs, a Rocketbook reusable notebook is ideal. Sketch visual drafts, jot down winning prompts, scan and archive everything to the cloud: you'll have your personal inspiration archive always at hand.
Step 4: Submit the request and wait for the results Press the "Send" button or the paper airplane icon to submit your prompt. Gemini will take a few seconds to process the request and generate the image.
Step 5: View and download the generated image Once processing is complete, you will see the generated image in the chat window. If you are satisfied with the result, you can download it by clicking on the image and selecting the download option. If the image doesn't exactly match what you had in mind, you can ask Gemini to make specific changes.
Step 6: Refine the image with follow-up requests You can improve the generated image by asking for specific modifications. For example, you can write: "Can you make it with warmer colors?" or "Add more people to the scene." Gemini will generate a new version of the image incorporating your requests.
According to a study by iapertutti.it, one of the most impressive features of these models is their ability to maintain subject consistency across successive edits. This means you can request changes to the image (like color or adding elements) while keeping the same subject and style as the original image.
Step 7: Edit an existing image Gemini can also edit existing images. Click the paperclip icon (or the "Add file" option) in the toolbar, upload an image from your device, and then write how you would like to modify it. You can ask to change the background, add elements, change the style, or alter colors.
Step 8: Experiment with different styles and formats Try specifying different artistic styles in your prompts, such as "watercolor", "photography", "anime style", "pixel art", or "pencil drawing". You can also request specific formats like "panorama", "vertical", or "square" depending on your needs.
Creative sessions can be long. To maintain focus while experimenting with styles and formats, a comfortable work environment is essential. An ergonomic mouse like the Logitech MX Master 3S reduces fatigue when navigating between different generated images, while noise-canceling headphones like the Sony WH-1000XM5 help you isolate yourself and enter a creative flow state.
How to use OpenAI's DALL-E
If you prefer to use DALL-E, OpenAI's image generation tool, the process is similar but requires an OpenAI account. If you have already followed our previous ChatGPT tutorial, access will be easier because you can use the same account.
- Visit the OpenAI website (openai.com) or go directly to ChatGPT (chat.openai.com)
- Log in with your account or create a new one
- Once inside ChatGPT, you can ask it to generate an image by writing a prompt like "Generate an image of..."
- You can also access DALL-E directly via labs.openai.com and use the dedicated interface
DALL-E is particularly powerful for generating photorealistic and detailed artistic images, but it requires a ChatGPT Plus subscription for unlimited use. For a more in-depth tutorial on DALL-E, Salvatore Aranzulla has created a detailed guide that also explains how to edit existing images with this tool.
Bonus: Tips for creating effective prompts
The quality of the generated images largely depends on how detailed and precise your prompts are. Here are some tips to get better results:
- Be specific and detailed: Instead of writing "a dog", specify "a Siberian Husky puppy playing in the snow with a red scarf".
- Describe the artistic style: Mention the style you want, such as "photorealistic", "oil painting", "anime style", "3D illustration", etc.
- Specify lighting and atmosphere: Terms like "soft light", "dramatic lighting", "golden sunset", or "morning fog" can significantly improve the result.
- Mention the perspective: "Close-up", "top-down view", "wide-angle", or "eye-level shot" help define how the image should be composed.
- Use artistic references: You can mention famous artists or well-known styles, such as "in the style of Van Gogh" or "like a Wes Anderson film".
Remember that these tools have ethical and legal limitations. They cannot generate explicit, violent, offensive content or content that violates copyright. Furthermore, AI-generated images may have inconsistencies or unrealistic elements, so always check them carefully before using them in professional contexts. As highlighted in a Datamasters article, all images generated by DALL-E 3 include an invisible watermark that identifies them as AI-created.
Creativity also needs moments of pause. To give your eyes a break from the screen after a long image generation session, treat yourself to an analog activity. Creating relaxing ambient lighting with a Philips Hue smart lighting kit can help your mind recharge and find new inspiration.
Conclusion: A New Way to Bring Ideas to Life
AI image generation is revolutionizing the way we create visual content. Whether you are a blogger, a marketer, an educator, or simply a creative person, these tools allow you to bring your ideas to life without the need for artistic skills or complex editing software.
Start with simple prompts and then experiment with increasingly detailed and creative descriptions. With a bit of practice, you'll discover that the real skill lies in "painting with words" – describing what you want to see so vividly that the AI can perfectly translate it into an image. As highlighted in a McKinsey 2025 report, image generation technologies are already transforming sectors like marketing, publishing, and design, with an estimated economic impact in the billions of dollars.
To discover other artificial intelligence applications that can simplify your work and stimulate your creativity, subscribe to our La Bussola dell'IA newsletter and visit our blog regularly.
To delve deeper into the incredible creative potential of AI beyond images, I recommend "AI 2041" by Kai-Fu Lee and Chen Qiufan. It's a fascinating exploration of how artificial intelligence will shape the future of creativity and society, perfect for someone like you who is already experiencing this revolution firsthand.