Artificial Intelligence

Generative AI can help doctors diagnose patients—but is it biased?

A new study by Annnenberg School for Communication professor Damon Centola tests if AI tools could help improve medical care without increasing bias. The findings suggest that “AI can meaningfully augment physician decision-making without introducing inequities in clinical decisions,” Centola says.

Why the true water footprint of AI is so elusive

Benjamin Lee of the School of Engineering and Applied Science says that net zero climate pledges can’t be achieved under current infrastructure investment plans for generative AI data centers.

Adventures in innovation: Penn Engineering startups lead Venture Lab Challenge

A trio of Penn Engineering startups took home more than $100,000 at the annual Venture Lab Startup Challenge. Sync Labs, whose AI assistant, Alice, is enhancing caregiving for seniors, claimed the Richard and Ellen Perlman Grand Prize, as well as three other awards. Quok.it, which helps customers leverage unused computing power around the world for tasks like AI training, earned the William G. Simpson and R. Drew Kistler Runner Up Prize.

Five ways gen AI is changing workplace identity

At this year’s SXSW Conference, Wharton’s Stefano Puntoni spoke with Google chief measurement strategist Neil Hoyne about the impact of generative AI on employees and the workplace.

Lovelace’s theory in practice: Our learnings from 2000 gen AI projects

Ethan Mollick of the Wharton School says that some business leaders may be missing out on the economic potential of AI.

When ChatGPT broke an entire field: An oral history

Christopher Callison-Burch of the School of Engineering and Applied Science shares his reaction to experimenting with the GPT-3 beta in 2020, the precursor to OpenAI’s ChatGPT generative AI model.

How MLR@Penn supports student AI researchers

Machine Learning Research @ Penn prepares undergraduates for research by discussing academic papers in small groups, much like book clubs would dissect a novel.

2 min. read

Can we still detect AI-generated content?

As models like GPT-4 and Claude get better at mimicking humans, researchers at the Wharton School offer a new way to test watermarking—the hidden markers used to identify machine-made text.

Tianwei Wu (left) and Liang Feng (right) demonstrating some of the apparatus used to develop the new, light-powered chip.

Penn engineers first to train AI at lightspeed

Penn engineers have developed the first photonic chip that reshapes how light behaves to carry out the nonlinear mathematics at the heart of modern AI while reducing energy use.