Incident 173: Unintended AI Response Causes Service Disruption
Recent incident #173 highlights the importance of responsible AI governance as an unintended response by an AI system caused a temporary service disruption. The AI model, designed for customer support, mistook a harmless statement as a request for account deletion, leading to the affected user's account being closed prematurely. This underscores the need for safe and secure AI systems that are equipped with effective guardrails. HISPI Project Cerebellum TAIM provides tools to Map, Measure, and Manage such incidents to prevent future harm.
Source
Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/173
Data source
Incident data is from the AI Incident Database (AIID).
When citing the database as a whole, please use:
McGregor, S. (2021) Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database. In Proceedings of the Thirty-Third Annual Conference on Innovative Applications of Artificial Intelligence (IAAI-21). Virtual Conference.
Pre-print on arXiv · Database snapshots & citation guide
We use weekly snapshots of the AIID for stable reference. For the official suggested citation of a specific incident, use the “Cite this incident” link on each incident page.