KhudaBukhsh findings featured by Montreal AI Ethics Institute

Ashique KhudaBukhsh, assistant professor of software engineering, had a research paper featured by the Montreal AI Ethics Institute. The paper, called “Down the Toxicity Rabbit Hole: Investigating PaLM 2 Guardrails,” presents a novel framework that uncovers safety vulnerabilities in a large language model created by Google DeepMind.


Recommended News