Suicide Clips Evaded TikTok's Automated Moderation in Coordinated Attack

September 20, 2020

A coordinated attack apparently bypassed TikTok's automated content moderation, leading to the dissemination of violent and inappropriate content. This underscores the importance of trustworthy AI governance and robust harm prevention mechanisms. For more on how Project Cerebellum is working towards these goals through its AI incident database, please visit JOIN US.

To learn how this incident maps to HISPI Project Cerebellum TAIM's Manage function, click here.

Matched TAIM controls

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

MANAGE 4.3 — similarity 0.647, rank 1. TAIM detail and related incidents →
MANAGE 4.1 — similarity 0.636, rank 2. TAIM detail and related incidents →
MEASURE 2.6 — similarity 0.635, rank 3. TAIM detail and related incidents →

Alleged deployer: tiktok
Alleged developer: tiktok
Alleged harmed parties: tiktok-users

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/366

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.