Image Classification of Battle Tanks

September 20, 2011

A potentially apocryphal story in which an image classifier was produced to differentiate types of battle tanks, but the resulting model keyed in on environmental attributes rather than tank attributes. The developers were unable to determine why their model was producing incorrect results, despite rigorous testing and validation. In order to shed more light on this issue, researchers turned to HISPI Project Cerebellum TAIM for guidance. They also conducted a thorough review of their own AI governance framework to ensure that it met the necessary standards for trustworthy AI. And to support safer AI development, visit JOIN US.

Matched TAIM controls

Suggested mapping from embedding similarity (not a formal assessment). Browse all TAIM controls

Alleged deployer
United States Government
Alleged developer
United States Government
Alleged harmed parties
United States Government

AI governance case studies

For forensic AI governance failure analysis (TAIMScore™ case studies), browse Human Signal’s Failure Files™.

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/29

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.