May 20, 2026Computer Vision, Generative AI

The Neural Turing Test: Can Classifiers Detect AI-Generated Images?

My thesis research centers on a deceptively simple question — can a neural network tell a real photo from a synthetic one? Here's what I've found so far, and why it's harder than it sounds.

Read more →

April 10, 2026LLMs, Bias

Quantifying Nationality-Based Compliance Bias in LLMs

Large language models don't treat everyone equally. My research at the Language GRASP Lab found systematic patterns in how open-source models respond to requests depending on perceived user nationality — and we can localize this in the network.

Read more →