On the Misuse Of Generative AI and What We Can Do About It

Talk by Erwin Quiring

2025-05-06

Location: TU Wien, FAV Hörsaal 3 Zemanek (Seminarraum Zemanek, Favoritenstraße 9-11, Erdgeschoß) (HEEG02)
Date/Time: 2025-06-02 17:00 ‒ 18:00

Abstract: The emergence of generative artificial intelligence (AI) has revolutionized content creation, enabling us to produce highly authentic digital media such as images, videos, texts, and music that is hard to distinguish from real media. Despite many positive applications, this also leads to considerable concerns on the misuse potential, such as recent incidents in elections or recent phishing attempts have effectively demonstrated.

In my talk, we revisit how AI is misused and how it might be misused in the future. Then, we dive into possible countermeasures to mitigate this misuse, including passive detection approaches and recent watermarking trends.

Bio: Erwin Quiring is a researcher at the Ruhr University Bochum, Germany. He received a Ph.D. in computer science from TU Braunschweig and has worked as postdoctoral researcher at ICSI @ UC Berkeley from 2023 to 2024. His research vision is to foster the trustworthiness of AI. In particular, he works on the reliability, efficiency, and security of AI. One of this main research areas is the detection of AI-generated content. Contact him at erwin.quiring@rub.de