A graphical summary in an article is one of the most useful tools for communicating hard-won research results. It makes findings more memorable, accessible, and engaging for everyone, from experts in the field and policymakers to curious readers outside the field.

But creating graphic briefs isn’t easy, and not everyone has access to a graphic designer or is comfortable with design software. Artificial intelligence tools can help fill that gap, if used thoughtfully.

Three tips to avoid AI image mistakes in science

Before AI, when designing graphical briefs, I spent a lot of time determining what to do and what not to do, what tools to use, and how to get started. Now I can use AI to dictate ideas through its speech-to-text features, vectorize sketches, optimize color palettes for accessibility, find references, brainstorm, and create custom tools that fit an exact workflow.

But with greater convenience comes greater responsibility. An AI-assisted chart has its name, not that of the AI. It is up to you, the user, to ensure that the image is scientifically accurate and does not plagiarize or infringe the work of others.

So how to create quality graphical summaries while integrating AI responsibly and effectively? These are some of the lessons I have learned.

Think before you start

First, check the target magazine’s guidelines on using AI. Some magazines prohibit AI-generated images and others require a clear statement.

Assuming the use of AI is allowed, the next step is to determine what you want to communicate. Are you illustrating a mechanism, process, experimental setup, workflow, comparison, or timeline? For one of my articles, the key message was how enzymes called deubiquitinases drive obesity-related liver pathology and where in the disease process inhibitors of the enzyme would stop the progression or development of the disease.¹. So, rather than mapping each individual molecular interaction, the graphical summary (which in this case was created without AI) was designed to show how these enzymes contribute to disease progression using a single representative pathway rather than the full network of interactions identified in the literature. The goal was to provide enough context to make the mechanism intuitive and engaging, while leaving room for readers to explore the molecular details of the entire study.

Next, think about your target readers. Are they doctors, industry professionals, policymakers, students, or experts in the field? Knowing your audience helps you determine the appropriate level of detail, terminology, and visual complexity. Academic readers might prefer a biochemical approach, while clinicians discuss clinical implications such as inflammation and insulin resistance. Make the summary relevant to your audience by focusing on the part of the research that matters to them.

‘Good design requires mastery’: scientific illustrators outline the future of AI

Finally, look for the literature. How do your graphical summaries present similar articles, both in content and composition? Which ones work well and why?

AI can help augment these brainstorming steps. Tools like ChatGPT (from OpenAI in San Francisco, California) and Claude (from Anthropic in San Francisco) can help distill the key message of the study, identify relevant articles, and allow you to better understand your readers. AI tools can even suggest pictorial representations of broad terms or concepts that you want to represent abstractly, such as multiomics or systems-level interactions.

Communicate with design

What creates an AI tool depends on how clearly you communicate your idea. And for that, you need a well-written message. Make intentional decisions about visual aspects such as:

Color: Aligning color choices with psychology makes the abstract more intuitive. Red and orange attract attention and are often perceived as negative or dangerous, so researchers frequently use “warm” colors like these to indicate key interactions or harmful elements such as cancer cells. “Cool” colors, on the other hand, tend to indicate normal or healthy conditions.

celebrating NatureCovers: past, present and future

But less is more and too much color (or too many shapes) can cause confusion. Try to create a limited, consistent, high-contrast palette.

Ask your AI tool: What color can I use to ensure color accessibility on a given background? Or, what colors should I assign if I want to show a particular pathway in a specific cell type?

Composition: Do you prefer a horizontal or vertical flow, a cyclical circuit or a branching path? Each one directs their gaze differently. A good composition guides the eye across the figure in the desired direction.

Ask your AI tool: Here is a graphical summary of my research. Can you suggest alternative compositions? Can you recommend an alternative text placement?