Resolving AI Incident #179: Unintended Model Drift
Recently, our team encountered an incident involving unintended model drift in Project Cerebellum's AI model. This instance underscores the importance of safeguards for AI to ensure safe and secure operations. The model was designed to predict stock market trends but started producing incorrect results, leading to potential financial harm. Through careful analysis, our team identified the root cause: lack of regular updates in the training data. By promptly addressing this issue, we prevented further potential harm and reinforced the need for continuous monitoring and maintenance in AI governance.
Source
Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/179
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.