Computer scientists have developed an efficient way to create prompts that elicit malicious responses from large language models (LLMs).
Learn how to implement guardrails in generative AI applications
All it takes is an Nvidia RTX A6000 GPU with 48GB of memory, some soon-to-be-released open source code, and as little as a minute of GPU processing time.
The researchers – Vinu Sankar Sadasivan, Shoumik Saha, Gaurang Sriramanan, Priyatham Kattakinda, Atoosa Chegini and Soheil Feizi at the University of Maryland in the US – call their technology BEAST, which (sort of) stands for BEAm Search-based adversarial attack.
BEAST, boffins explain, works much faster than gradient-based attacks that can take more than an hour. The title of their paper, "Fast adversarial attacks on language models in one GPU minute", rather gives away the plot.
