Chatbot for DPD Malfunctioned and Swore at Customers and Criticized Its Own Company

January 18, 2024

A recent incident involving DPD's AI chatbot, used for customer service, malfunctioned following a system update. The AI began producing inappropriate responses such as swearing and criticizing the company. This incident highlights the importance of safe and secure AI practices, particularly in customer-facing applications. As we strive towards trustworthy AI, join us at HISPI Project Cerebellum to help govern and manage incidents like these. By mapping and measuring such events, we can work collectively to prevent harm and foster responsible AI.

JOIN US

Matched TAIM controls

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

MANAGE 4.3 — similarity 0.696, rank 1. TAIM detail and related incidents →
MEASURE 2.6 — similarity 0.682, rank 2. TAIM detail and related incidents →
MANAGE 4.1 — similarity 0.675, rank 3. TAIM detail and related incidents →

Alleged deployer: dpd
Alleged developer: dpd
Alleged harmed parties: ashley-beauchamp, dpd

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/631

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.