AI Incident Under-Reporting: Behavioral, Institutional, and Technological Drivers Across Critical Sectors

While the literature acknowledges under-reporting as a persistent challenge (Fernando et al., 2023; Listiowati et al., 2024), there has been little cross-sectoral, multi-method inquiry into its root causes in the context of AI. This research would combine qualitative interviews, behavioral experiments, and large-scale survey data to map the interplay of institutional incentives, fear of blame, knowledge gaps, and technological barriers to reporting. It would further develop and pilot interventions, such as anonymous reporting channels, gamified incentives, and automated error detection nudges. The novelty lies in its comparative, empirical focus across multiple sectors (as highlighted by Agarwal & Nene, 2025 for telecom and Dillon et al., 2024 for aviation), and in its aim to move beyond policy recommendations toward empirically validated solutions to the under-reporting problem.

References:

1. Are Incident Reporting Systems in Healthcare Systems a Requirement for Improving Patient Safety? A Review. G. Fernando, T. Bandara, M. Purva (2023). International Journal of Health Systems Resilience.
2. Patient Safety Incident Reporting Challenges in Indonesian Private Hospitals. E. Listiowati, M. Arini, M. Nurmansyah, Emma Rachmawati, Agus Samsudin (2024). Global Medical & Health Communication (GMHC).
3. Incorporating AI Incident Reporting into Telecommunications Law and Policy: Insights from India. Avinash Agarwal, M. Nene (2025). arXiv.org.
4. How AI Can Help Learn Lessons from Incident Reporting Systems. Robin Dillon, Peter Madsen, Brian Holland, Danniel Cao (2024). IEEE Aerospace Conference.