Indirect Prompt Injection Into LLMs Using Images and Sounds
Offered By: Black Hat via YouTube
Course Description
Overview
Explore the security implications of multi-modal Large Language Models (LLMs) in this 28-minute Black Hat conference talk. Delve into the potential vulnerabilities of advanced AI systems like Bard and future iterations of ChatGPT that process various input types including text, audio, and images. Discover how attackers can exploit these models through indirect prompt and instruction injection using adversarial perturbations in images and audio recordings. Learn about the techniques used to manipulate unmodified, benign multi-modal LLMs into outputting attacker-chosen text or following malicious instructions. Gain insights from researchers Ben Nassi and Eugene Bagdasaryan on this emerging threat to AI security and its potential impact on the future of language models.
Syllabus
Indirect Prompt Injection Into LLMs Using Images and Sounds
Taught by
Black Hat
Related Courses
AI CTF Solutions - DEFCon31 Hackathon and Kaggle CompetitionRob Mulla via YouTube Indirect Prompt Injections in the Wild - Real World Exploits and Mitigations
Ekoparty Security Conference via YouTube Hacking Neural Networks - Introduction and Current Techniques
media.ccc.de via YouTube The Curious Case of the Rogue SOAR - Vulnerabilities and Exploits in Security Automation
nullcon via YouTube Mastering Large Language Model Evaluations - Techniques for Ensuring Generative AI Reliability
Data Science Dojo via YouTube