Konrad Kollnig
Associate Professor, Law & Tech Lab
Maastricht University
Konrad Kollnig is an associate professor at Maastricht University's Law & Tech Lab in Maastricht, the Netherlands, a computer scientist in a law faculty who tests what apps and AI systems do and whether the laws that govern them work. His team's audit of shopping advice from ChatGPT, Gemini and Google's AI Overviews found that the products recommended and the sources shown change from one request to the next.
Auditing Chatbot Shopping Advice
"With companies like OpenAI and Google monetising their AI through advertising, this raises difficult questions about the bias and impartiality of such advice," Kollnig and his co-authors wrote in "If I Had to Buy Just ONE: Galaxy S26 Ultra": Auditing AI-Generated Product Recommendations, released in September 2026 with lead author Lucas G. Uberti-Bona Marin and researchers from Utrecht University and the University of Zurich. The team compiled 2,528 real commercial-advice queries and evaluated 1,536 responses from ChatGPT and Gemini, through both their chat apps and their APIs, and from AI Overviews in Google Search.
ChatGPT expressed a first-person product preference in 79% of its product-recommending answers, against 7% for Gemini and 2% for AI Overviews, and the products it recommended often changed across repeated requests. Sources varied even more: for the same query, the ChatGPT and Gemini apps shared only 5.4% of displayed domains on average, and the APIs gave a different picture from the apps consumers use. Independent audits, the authors concluded, "should therefore account for repeated responses, consumer-facing conditions, and the source layer being observed."
Tracking the App Economy
"Most apps quietly track us," Kollnig writes on his personal site. To make that visible he built TrackerControl, an open-source Android tool based on an analysis of some two million apps and downloaded more than 200,000 times, and DuckDuckGo adopted its blocking approach for its App Tracking Protection. His book The App Economy, on how Apple and Google use their dominance in that market, was published open access by Bristol University Press in 2026.
At Maastricht he leads Comp4Good, an interdisciplinary team focused on AI auditing, holds the RegTech4AI fellowship to make the EU AI Act and GDPR work in practice, and co-leads REGULAIRE, a European doctoral network that helps regulators keep pace with technology. He also serves as chief technology officer of the Centre for Digital Habits, which builds open-source tools that give young people more control over social media.
From Aachen to Oxford and Maastricht
Kollnig studied computer science and mathematics at RWTH Aachen University in Germany, then took a master's degree in computer science and a PhD in computer science and law at the University of Oxford, where he was a DeepMind Scholar. His thesis won the Council of Europe's Stefano Rodotà Award in 2024 for the best PhD thesis in data protection. He joined Maastricht as an assistant professor in 2023 and became an associate professor in 2026.