Celebrities' Deepfake Voices Abused with Malicious Intent

January 30, 2023

Celebrity voices were manipulated using deepfake technology for malicious purposes, such as impersonation or slander, and disseminated on social media platforms like 4chan and Reddit.

Learn how Project Cerebellum is working towards safeguarding AI with our Harm Prevention Initiative. Join us in Governance of the HISPI Project Cerebellum TAIM to help prevent incidents like these and promote safe and secure AI practices. JOIN US

Matched TAIM controls

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

MAP 1.6 — similarity 0.612, rank 1. TAIM detail and related incidents →
MEASURE 2.10 — similarity 0.606, rank 2. TAIM detail and related incidents →
MAP 4.1 — similarity 0.599, rank 3. TAIM detail and related incidents →

Alleged deployer: reddit-users, elevenlabs-users, 4chan-users
Alleged developer: elevenlabs
Alleged harmed parties: public-figures, celebrities

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/508

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.