News

All News

Xiang Cheng: Using Geometry to Push AI’s Boundaries
9/19/24 Pratt School of Engineering

Xiang Cheng: Using Geometry to Push AI’s Boundaries

New faculty member Xiang Cheng exploits geometric patterns in data and algorithms to find better ways to train advanced AI programs

Shaundra B. Daily of Duke University
Duke Trailblazer: Shaundra Daily
9/9/24 Duke Centennial

Duke Trailblazer: Shaundra Daily

For Shaundra Daily, helping to create a more equitable STEM education system is an important part of her work

Tulip Garden
“Model Collapse” Threatens to Kill Progress on Generative AIs
9/4/24 Freethink

“Model Collapse” Threatens to Kill Progress on Generative AIs

Emily Wenger explains that if you feed an AI its own dog images enough times, errors in those images will eventually prevent it from being able to produce images that look like dogs at all.

Pratt School of Engineering
AI Learning From Its Own Nonsense Might Just Self-Destruct, Experts Warn
7/26/24 Gizmodo

AI Learning From Its Own Nonsense Might Just Self-Destruct, Experts Warn

Emily Wenger emphasizes the risks associated with recursive self-improvement in AI models. She highlights that without careful oversight, these systems might amplify their own mistakes, leading to unpredictable and potentially harmful outcomes.

Hudson Hall at Duke university
China Is Closing the AI Gap With the United States
7/25/24 The New York Times

China Is Closing the AI Gap With the United States

Yiran Chen discusses China's initiatives to bridge the gap in AI development and the influence of open-source AI models on the US-China competition.

Shyam Venkatasubramanian
Revolutionizing the Abilities of Adaptive Radar With AI
7/19/24 Pratt School of Engineering

Revolutionizing the Abilities of Adaptive Radar With AI

AI approaches and an enormous open-source dataset could spark rapid advancements in adaptive radar systems similar to those seen in computer vision over the past two decades