Experiment shows AI models have surprising preference for the color blue

A novel experiment revealed that AI language models demonstrate distinct color preferences across different systems and prompts. In the experiment, a researcher asked each model to select favorite colors and provide specific hex RGB color codes, revealing intriguing patterns in their responses.

Key findings: Large language models show a clear tendency toward blue hues, with indigo (#4B0082) emerging as a particularly common choice across multiple systems.

• When prompted for color preferences, most AI models gravitated toward various shades of blue, ranging from deep indigo to bright azure
• The Llama-3.2-vision-90b-instruct model consistently selected black, standing out from the general blue trend
• Results showed significant variation depending on how questions were phrased, highlighting the importance of prompt engineering in AI interactions

Model-specific patterns: Different AI systems demonstrated unique and consistent color preferences, suggesting underlying patterns in their training or decision-making processes.

• Deepseek models regularly chose indigo, teal, or azure shades
• Gemini displayed a clear preference for teal and sea green colors
• Opus and Mistral models consistently selected intense blue hues
• These patterns remained relatively stable across multiple testing sessions

Technical methodology: The investigation utilized the lmsys platform with default parameters to ensure consistency across testing.

• Researchers employed various prompt formats to elicit color preferences
• All responses were collected using standardized hex RGB color codes for precise comparison
• Testing maintained consistent parameters across all models to ensure reliable results

Research implications: The study raises important questions about how AI systems process and respond to subjective preferences.

• The consistent patterns suggest that AI models may develop inherent biases during training
• Color preferences could be influenced by the frequency of certain colors in training data
• The variation in responses based on prompt wording indicates the complexity of AI decision-making processes

Looking deeper: Understanding why AI models make specific color choices could provide valuable insights into machine learning processes and decision-making mechanisms, while also highlighting the importance of careful prompt engineering in AI interactions.