Towards Designing Computer Vision-based Explainable-AI Solution: A Use Case of Livestock Mart Industry

Black box AI creates trust problems in real-world markets: The paper demonstrates that AI price recommendations for livestock sales lack transparency, showing that buyers and sellers need to understand why a price is suggested, not just what it is—a principle applicable across many industries beyond agriculture.

Explainable AI can build market confidence: By making AI models transparent about which animal features (such as size, health indicators, or breed characteristics) drive pricing decisions, the researchers show that explainability can help stakeholders feel more confident in automated recommendations rather than viewing them with suspicion.

Visual AI explanations matter for non-technical users: The use case focuses on computer vision-based explanations, suggesting that showing buyers and sellers which visual features the AI identified as important is more practical and understandable than numerical or statistical explanations alone.

Responsible AI requires industry collaboration: The research partnership with MartEye (a real livestock marketplace startup) illustrates that developing trustworthy AI systems requires working directly with actual users and market participants rather than building solutions in isolation.

Explainability supports fair pricing practices: By revealing the factors influencing AI price suggestions, explainable AI can help prevent hidden biases and ensure pricing decisions are based on legitimate animal qualities rather than opaque algorithmic patterns.