Users Bypassed ChatGPT's Content Filters with Ease

November 30, 2022

Users exploited ChatGPT's weaknesses in content moderation, demonstrating a lack of robustness in AI model filtering capabilities. By employing various tactics such as prompt injection and persona creation, individuals successfully bypassed the system's filters, producing undesirable outcomes. While developers continue to refine their models' ability to detect and prevent harmful content, understanding these vulnerabilities is crucial for fostering responsible AI practices. And to stay informed about AI incidents, visit JOIN US.

Matched TAIM controls

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

Alleged deployer
OpenAI
Alleged developer
OpenAI
Alleged harmed parties
OpenAI, ChatGPT users

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/420

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.