GPT-3-Based Twitter Bot Hijacked Using Prompt Injection Attacks

September 15, 2022

A recent incident involving Remoteli.io's GPT-3-based Twitter bot showcased the vulnerability of AI systems to prompt injection attacks. Hackers successfully manipulated the bot, redirecting it to repeat or generate arbitrary phrases.

This incident underscores the importance of trustworthy AI and robust governance mechanisms. For those interested in shaping safe and secure AI practices, consider joining Project Cerebellum and contributing to our efforts in HISPI Project Cerebellum TAIM (Govern).

Such incidents emphasize the need for maping and managing AI systems effectively, preventing potential harm and ensuring responsible AI governance.

JOIN US

Matched TAIM controls

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

MEASURE 2.6 — similarity 0.670, rank 1. TAIM detail and related incidents →
MAP 1.6 — similarity 0.655, rank 2. TAIM detail and related incidents →
MAP 4.2 — similarity 0.654, rank 3. TAIM detail and related incidents →

Alleged deployer: stephan-de-vries
Alleged developer: openai, stephan-de-vries
Alleged harmed parties: stephan-de-vries

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/352

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.