How to Create AI Images: A Complete Guide with Expert Advice

A picture says more than a thousand words. But with generative AI, you have to start with words to get that picture.

Generative artificial intelligence technology has exploded in recent years, giving rise to many creative services that can generate entire images and video clips. This is the “generative” aspect of genetic AI – it can seemingly generate content from just a simple text prompt. Of course, the actual process is more complicated.

AI image generators rely on a database of existing content (text, images, videos, etc.) and/or search the open Internet for information to create new images to fulfill your request. They allow anyone with an account and a request to create almost any type of image (usually within reason).

There are many AI imaging services available, but not all of them provide high-quality results or adequately protect your data. They also differ in price and the features they offer. While AI generators can be exciting to explore, getting started can quickly become overwhelming.

This is where this guide comes in handy. I tested and reviewed several AI image generators for CNET and explained how the new technology impacts digital creators. These services are certainly interesting and serve their purpose, but they require a learning curve. Although they are not good enough to replace human creators, they offer a range of creative possibilities.

Here’s how you can create custom AI images.

Step 1: Choose the best AI service for your project

I have hands-on experience with many AI image generators and creative programs. Which program is best for you depends on your project, your budget and your level of experience. You can check out our full reviews and rankings, but these are my three program recommendations.

What is the best AI image generator for you?

The CNET Editor’s Choice is Dall-E 3 from OpenAI, the company behind ChatGPT. Dall-E features impressive prompt adherence and handles even more complex prompts with ease. The conversation flow – meaning you can follow simply by typing in the chat window – allows for easy editing after generation. Dall-E is also one of the more creative programs we tested. You’ll need a ChatGPT Plus account to access the program – $20 per month – and note that OpenAI may use your content for model training. However, you can submit a request to stop training your data.

If you are looking for a free program, I recommend Leonardo AI. Canva recently acquired it, but you can still use it on their website. Leonardo offers a robust free plan with more editing tools behind the paywall and multiple artistic styles to choose from. It also offers a unique quick improvement tool that allows you to rewrite a simple idea and add details that lead to beautiful end results.

If your budget is really tight and you have no creative experience, check out Canva’s Magic Media generator. It is extremely intuitive and easy to navigate, which is ideal for beginners and non-creatives. Canva offers you 50 free image generation credits (that’s lifetime credits – you’ll need to upgrade to the paid Pro plan for more credits after you use up your free tier) and automatically excludes you from training its AI on your content.

Once you’ve selected your program, create an account and go to the Command Prompt window.

Step 2: Write your image prompt

Your best chance of success starts with a well-written prompt. Each prompt should include character basics, setting, other elements, and dimensions of the image. But to really hone in on the image you want, add descriptions of the artistic style you want to emulate, the aesthetic, the appeal, and the specific colors.

AI image generators are rightly subject to limitations in terms of generation options. This is to prevent people from misusing these services to create illegal content and potentially dangerous or offensive content. For example, Dall-E 3 refuses to create images that mention public figures such as President Biden or Taylor Swift by name. Before generating, be sure to check your service’s security policies.

Prompt engineering – the process of writing the right prompt to get the desired results – is something you get better at with practice. It also varies between individual image generators. For example, Canva tends to perform better with simple prompts, while Dall-E recommends longer, more detailed prompts. The best way to get to know it is to play around with your program. You can also find more tips and tricks in my complete guide to image prompt engineering. Once you get the best prompt, you can continue generating it!

Step 3: Edit your images as needed

Even with the best generator, you’ll likely need to edit your images after they’re created. Depending on the service you use, you may have access to editing tools or the ability to request changes. Be careful when editing your images too much. I’ve found that services can only handle so many changes before the image becomes too wonky and ultimately unusable.

Post-generation editing tools are an area where AI services tend to shine or fail. If your service doesn’t offer them (or they don’t work particularly well), try editing your prompt instead of the image. In my experience, this gives you a better chance of success than trying to edit existing images incrementally.

Step 4: Give your images proper attribution

It is extremely important to label your images as AI-generated content. As AI image services improve, it is important that all creators clearly identify the origin of an image, particularly to distinguish it from human-made art and photography.

Some services automatically add a watermark or other marking indicating that it is AI generated, but too many do not, such as Dall-E, Adobe Firefly, Midjourney, Canva and Leonardo. If you share your image online, be sure to credit the AI ​​service you use in the caption, credit, and/or alt text. You can also add a disclaimer to your posts when sharing them on social media – Meta might flag them as AI content on Instagram and Facebook, but adding a line to your headline will help avoid confusion.

FAQ and other considerations

At the time of publication, using AI image generators is legal. However, when creating AI images, you should consider important legal and ethical considerations. Here are a few quick, frequently asked questions.

AI image services like Midjourney and Dall-E rely on existing content to create images. This content, either in a database or on the open web, is typically created by humans. Therefore, some creators and artists have serious concerns about the design and training of these programs. This has led to industry turmoil and some notable lawsuits, including an ongoing case against Stable Diffusion, an AI model used by popular generators.

If you’re concerned about using a service that could potentially mistreat creators, choose a service that clearly discloses how it trains its model. Some AI services rely on public databases such as Adobe Stock and Getty Images, where creators sign agreements with the companies to have their works included in these collections for specific uses. Adobe Firefly, for example, says it only uses content from its public Adobe Stock databases and images whose copyrights have expired. It also doesn’t train on your content.

As these AI services continue to develop, the associated legal and ethical principles will also change. If you want to create AI art on a regular basis, make sure you stay up to date with the latest requirements, which we cover in our AI Atlas guide.