Predicting Space Tourism Demand Using Explainable AI

The framework prioritizes transparency in AI decision-making: The researchers developed SpaceNet following National Institute of Standards and Technology guidelines for explainable AI, demonstrating that predictions about consumer behavior can be made interpretable rather than opaque "black boxes"—an important principle for responsible AI systems that affect business and consumer decisions.

Price and safety concerns are primary decision factors: The study identified that travel price, fatality probability, age, annual income, and gender are the most important features influencing whether someone wants to travel to space, suggesting that responsible marketing and safety communication should emphasize these factors rather than other variables.

The model achieves strong but not perfect accuracy: With an average ROC-AUC of 0.82 ± 0.088, SpaceNet demonstrates reliable performance, though the margin of error (±0.088) indicates the model has limitations—a responsible finding that acknowledges the technology cannot predict individual preferences with certainty and should not be used as the sole basis for business decisions.

Individual-level explanations enable personalized insights: Beyond aggregate predictions, the explainable AI approach provides interpretations for individual travel-type decisions, allowing businesses to understand the specific factors driving each person's interest and potentially tailor communications responsibly rather than using one-size-fits-all marketing.

Data collection was geographically and demographically diverse: The researchers collected 1,860 data points across multiple US states and cities with varying ages, suggesting efforts to avoid bias—though responsible implementation would require ongoing monitoring to ensure the model performs equitably across different demographic groups.