Automated Private Enforcement: Evidence from the Google Fonts Case

Can an automated private enforcement via AI-powered enforcement bots substitute for class action? Our new Working Paper suggests it can and might lead to a major disruption within the legal system. We track 1.5 million websites across 32 European countries to measure the impact of an automated private enforcement campaign in Austria in the wake of a court ruling holding that the dynamic embedding of Google Fonts violated the GDPR.

• Within 3 months alleged non-compliance in Austria fell by 22.7 percentage points, a 50% reduction. Both specific and general deterrence matter.
• When the marginal cost of sending a legal threat falls near zero, even minor violations can trigger large-scale enforcement.
• The reach of AI-driven enforcement is broader than online privacy. Similar bots now flag illegal logging via drones, detect deforestation from satellites, and mobilize citizens to report consumer harms.
• Regulators will have to adapt. With marginal enforcement costs driven toward zero, rules that were once under enforced can flip into over enforcement, raising fresh questions about how much—and how aggressively—we want the law applied.

Read more in their paper.

Professor Dr Alexander Stremitzer (ETH)

Before joining ETH Zurich in May 2018 as the Professor of Law, Economics, and Business, Alexander Stremitzer was a Full Professor of Law at UCLA from 2015 to 2018 and an Assistant Professor of Law at UCLA from 2011 to 2015. Prior to that, he was Assistant Professor of Economics at the University of Bonn. He was a visiting professor at Yale Law School, at Yale University’s economics department, at Northwestern Pritzker School of Law, at UCLA Law School, and at Harvard Law School. He also spent extended research visits at NYU Law School, ETH Zurich and Columbia University’s Center for Contracts and Economic Organization. In 2020, he was elected to the Board of Directors of the American Law and Economics Association, and in 2024, he was elected to the Board of Directors of the European Association of Law & Economics.

Alexander Stremitzer’s research and teaching interests include theoretical and experimental law and economics, contracts, business bankruptcy, and contract design. In addition to works in German, his recent scholarly work in English has been published in several journals, including the Journal of Legal Studies, the Journal of Law, Economics, & Organization, the Journal of Law & Economics, the American Review of Law & Economics, Games & Economic Behaviour, the Journal of Institutional and Theoretical Economics, the University of Chicago Law Review Online, the Harvard Journal of Law and Technology, and the Yale Law Journal. Furthermore, Alexander conducts lab, internet, and field experiments on legal and social institutions. He is also interested in the effects of new technologies on legal policy, regulation, and private ordering.

Dr Jakob Merane (ETH) 

Jakob is a postdoctoral researcher at the Center for Law & Economics at ETH Zurich and at the Faculty of Law at the University of Lausanne. He holds a PhD. from ETH Zurich and earned both his Bachelor and Master of Law (summa cum laude) from the University of Basel, with study visits at Humboldt-University of Berlin and University of Zurich. He concluded the Zurich bar exam with the highest honours and holds a Certificate of Advanced Studies in Applied Statistics from ETH Zurich. He is currently a Visiting Researcher at the Max Planck Institute for Research on Collective Goods and was a Visiting Researcher at Harvard Law School in 2023.

Jakob’s research explores the application of artificial intelligence in legal enforcement and legal technology. He draws on a multidisciplinary toolkit of computational, empirical, and experimental methods. His doctoral dissertation specifically examined the enforcement of small claims by using an automated web crawler and a machine learning pipeline to systematically detect data protection violations. Currently, his research centres on evaluating the capabilities and limitations of Large Language Models (LLMs) in legal contexts. Specifically, Jakob assesses their effectiveness in managing complex legal tasks, such as translating multilingual legal texts (SwiLTra), answering sophisticated legal exam questions (external pageLEXam), and examining their adaptability and cultural alignment with the Swiss legal context. He also develops Retrieval-Augmented Generation (RAG)-based tools for judicial institutions.