BlenderBot 3 Cited Dutch Politician as a Terrorist

August 25, 2022

The incident highlights the need for developers to thoroughly test and validate their AI models against sensitive topics, ensuring that they can distinguish between fact-based information and potentially inflammatory or misleading data. By reporting such incidents, users help drive improvements in AI safety and accountability, ultimately contributing to a safer online environment where accurate information is prioritized over misguided associations. We will continue to monitor BlenderBot 3's performance and provide updates on any necessary corrections or revisions, as well as offer guidance for users who may have been affected by this incident JOIN US.

Matched TAIM controls

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

Alleged deployer
Meta
Alleged developer
Meta
Alleged harmed parties
Marietje Schaake

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/313

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.