Kavita completed her PhD at the University of Texas at San Antonio (UTSA) and, during her 5th year, became involved in an exciting collaboration between TU Darmstadt and the System Security Lab. This collaboration eventually led Kavita to join us in the lab to work on an important research project. In 2022, she officially joined the System Security Lab as a Postdoc and took on the leadership of our AI team.

In addition to her leadership role, Kavita works on her own scientific projects, where she explores new approaches and research directions. One standout project is DEMASQ: Unmasking the ChatGPT Wordsmith, where she developed an innovative approach to detect AI-generated texts. In this project, Kavita designed an energy-based model that uses the Doppler effect to model the dynamics of communication between humans and machines. This allowed her to create a system that distinguishes between human-written and ChatGPT-generated content.

Kavita firmly believes that Physics-Augmented AI plays a crucial role in driving innovation. She believes that when people can look behind this technology, it will lead to various solution approaches for challenges in Large Language Models (LLM) and mathematics, helping to solve fundamental problems in machine learning and AI, and open up new possibilities.

In our interview, Kavita shared insights into her journey from PhD to Postdoc at the System Security Lab, discussed the challenges and opportunities in AI research, and spoke about her work in Physics-Augmented AI and her passion for developing innovative technologies.

We are proud to have Kavita as part of our team and are excited to share her valuable insights with you!

The System Security Lab at TU Darmstadt provides a dynamic and collaborative environment that fosters groundbreaking research. With access to top-tier resources and a vibrant academic community, researchers are empowered to push the boundaries of innovation in security and AI.