Explainable AI (XAI) in High-Stakes Domains: A Review of Techniques, Challenges, and Applications
Abstract
As Artificial Intelligence systems are increasingly deployed in high-stakes domains such as healthcare, finance, and law, the need for transparency and trust has become paramount. This paper provides a comprehensive review of Explainable AI (XAI) techniques, including model-agnostic and model-specific approaches. It evaluates methods such as SHAP, LIME, and attention mechanisms in improving interpretability. The study discusses the trade-offs between model performance and explainability, along with challenges in standardization and user trust. Real-world applications and regulatory implications are also examined. The paper concludes by identifying research gaps and proposing future directions for human-centric AI systems.
References
Szeliski, R. (2010). Computer vision: Algorithms and applications. Springer.
Kaur, K., & Sood, S. K. (2017). Machine learning-based anomaly detection in IoT systems. Journal of Network and Computer Applications, 89, 1–15.
Al-Fuqaha, A., Guizani, M., Mohammadi, M., & Aledhari, M. (2015). Internet of Things: A survey on enabling technologies. IEEE Communications Surveys & Tutorials, 17(4), 2347–2376.
Singh, B., Anand, A., Prabhat, S., & Ranjan, P. (2025). Threat Onboarding and Response (TOR): Automating Cybersecurity in Enterprise Networks.
Anand, A., Singh, B., & Prabhat, S. (2022). Real-Time Network Monitoring and Incident Response with AI-Driven Automation Data Center and WAN Transformation. Available at SSRN 5577033.
Anand, A., Singh, B., Khemka, S., Banerjee, B., Bhatia, V. S., & Ranjan, P. (2025, October). Malware Classification using Diluted Convolutional Neural Network with Fast Gradient Sign Method. In 2025 2nd International Conference on Software, Systems and Information Technology (SSITCON) (pp. 1-5). IEEE.
Singh, B., Bhatia, V., Anand, A., Edamadaka, G., & Sudhakar, M. (2025, June). Intent-based Software Defined Networking for Flexible Network Management in Heterogeneous Environments. In 4th Conference on SGCNSP-2025.
Whig, P., Sharma, P., Elngar, A. A., & Silva, N. (2025). Artificial intelligence and machine learning for advanced data-driven systems. CRC Press.
Whig, P., Sharma, P., & Yathiraju, N. (2025). Cutting-edge solutions for advancing sustainable development: Exploring technological horizons for sustainability (Part 1). Bentham Science Publishers.
Whig, P., Yathiraju, N., & Sharma, P. (2025). Cutting-edge solutions for advancing sustainable development: Exploring technological horizons for sustainability (Part 2). Bentham Science Publishers.
Whig, P., & Elngar, A. (2025). AI innovations for transforming food production. IGI Global.
Whig, P., & Elngar, A. (2025). Modernizing the food industry: AI-powered infrastructure, security, and supply chain innovation. IGI Global.
Whig, P., Sharma, P., Elngar, A. A., & Silva, N. (2025). Quantum learning: Bridging artificial intelligence, quantum computing, and data science in education. CRC Press.
Rather, R., Vats, H., Sharma, S., & Whig, P. (2025). AI and ML applications in data-informed leadership: Transforming higher education. In Advances in data science driven technologies (Vol. 5, pp. 36–53). Bentham Science.
Whig, P., Silva, N., Elngar, A. A., Aneja, N., & Sharma, P. (2025). Sustainable development through machine learning, AI and IoT: Proceedings of ICSD 2025 (Part I). Springer.
Whig, P., Silva, N., Elngar, A. A., Aneja, N., & Sharma, P. (2025). Sustainable development through machine learning, AI and IoT: Proceedings of ICSD 2025 (Part II). Springer.
Whig, P., et al. (2025). Shield-Pay: Secure healthcare encryption and integrated ledger for data and payments authentication yield (Patent application). Government of India.
