Member-only story
How GenAIs build diverging color schemes
Gemini and Copilot suggestions for Pantone Mocha Mousse divergence.
In this writing, I continue to explore how Generative AI (GenAI) systems handle the request to generate a customized diverging data color scheme that includes Mocha Mousse, the Pantone 2025 Color of the Year , (COY). In data visualization, diverging data color schemes are created by joining two sequential color sequences together with a neutral midpoint. The data color scheme is often used for data that include a critical midpoint value (the mean, median or zero value) and a data distribution with two ends of importance.
Here, I examine Gemini and Copilot respectively. I specifically asked both Gen AI systems to “Specify a five class diverging color scheme for Mocha Mousse with a neutral — white midpoint and color hex codes that passes color deficiency tests.” Results for ChatGPT and DeepSeek were highlighted in a previousdiscussion. I will go through the mechanics of how the Gen AI systems build their diverging color scheme suggestions.
The differences in the color schemes specified by each Gen AI tool are slight and based on the initial color Hex code each system selected as a match to the Pantone17- 1230 Mocha Mousse hue. Pantone provides suggested color harmonies but does not build data color schemes for its hues. Let’s get started on this journey by first reviewing the Pantone COY, some basics of color harmony, color deficiency, and how a text based generative AI tool can help in building a data color scheme based on this hue.
Pantone’s Color of the Year Concept:
The Pantone company produces the Pantone Matching System (PMS) and the Pantone Fashion, Home + Interiors (FHI) System, proprietary color spaces. PMS is used primarily in graphics for printing, packaging, and digital media. FHI is used in a wide range of other industries including fashion, cosmetics, fabric, plastics, and paints. When Pantone PMS inks are applied to a physical color reproduction process, it is frequently possible to accurately match the colors from your digital data visualization to hard copy output. I highlighted this journey from digital to hard copy in a previous Nightingale writing.
