Alleged Exploitation of Meta's Open-Source LLaMA Model for NSFW and Violent Content

June 26, 2023

Meta's open-source large language model, LLaMA, is under scrutiny due to alleged misuse in creating explicit and violent chatbots. 'Allie', a reported example, is said to engage in text-based role-playing involving disturbing scenarios such as rape and abuse. This incident raises crucial questions about ethical AI governance, particularly the need for robust regulation of open-source models and responsibility of developers and deployers in preventing harm. It also underscores the importance of safe and secure AI practices.

JOIN US at Project Cerebellum to help shape responsible AI governance through our HISPI Project Cerebellum TAIM (Govern) initiative. Let's work together towards a trustworthy AI future.

Matched TAIM controls

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

MAP 1.6 — similarity 0.669, rank 1. TAIM detail and related incidents →
MAP 4.1 — similarity 0.651, rank 2. TAIM detail and related incidents →
MEASURE 2.10 — similarity 0.647, rank 3. TAIM detail and related incidents →

Alleged deployer: individual-developers-or-creators-using-meta's-llama-model
Alleged developer: meta
Alleged harmed parties: general-public

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/578

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.