Computer Science

AI security

As AI gets more adept at synthesizing information and producing humanlike responses, many are concerned that malicious actors may use this technology in dangerous ways. Ph.D. candidate Alex Robey safeguards AI systems against malicious tampering.

Nathi Magubane

Real or fake text? We can learn to spot the difference

Penn computer scientists prove that people can be trained to tell the difference between AI-generated and human-written text. Their new paper debuts the results of the largest-ever human study on AI detection.

From Penn Engineering Today



In the News


Technical.ly Philly

Penn professor on gen AI’s rapacious use of energy: ‘One of the defining challenges of my career’

Benjamin Lee of the School of Engineering and Applied Science says that hardware and infrastructure costs are growing at high rates for generative AI.

FULL STORY →



Big Think

Can we stop AI hallucinations? And do we even want to?

Chris Callison-Burch of the School of Engineering and Applied Science says that auto-regressive generation can make it difficult for language learning models to perform fact-based or symbolic reasoning.

FULL STORY →



CNET

How the solar eclipse will affect solar panels and the grid

Benjamin Lee of the School of Engineering and Applied Science says that the electrical grid will have to figure out how to match supply and demand during brief windows where the energy source goes away.

FULL STORY →



CNBC

Students can soon major in AI at this Ivy League university—it’ll prepare them for ‘jobs that don’t yet exist’

The Raj and Neera Singh Program in Artificial Intelligence at Penn will be the first AI undergraduate engineering major at an Ivy League school, led by George Pappas of the School of Engineering and Applied Science.

FULL STORY →



NPR

Looking back at the transformative first year of ChatGPT

Michael Kearns of the School of Engineering and Applied Science says that ChatGPT could be remembered one day as being as important as the invention of the iPhone, or even the internet itself.

FULL STORY →



Tech Crunch

As OpenAI’s multimodal API launches broadly, research shows it’s still flawed

Chris Callison-Burch of the School of Engineering and Applied Science and Ph.D. student Alyssa Hwang provide their early impressions of GPT-4 with vision.

FULL STORY →