CITP’s mission is to understand and improve the relationship between technology and society. CITP aims to harness Princeton’s unique strengths in computer science and engineering, policy, the social sciences, and humanities to help governments, institutions, and individuals leverage the benefits of technology while mitigating its potential to do harm. By tightly integrating basic research, applied research, and societal engagement, we enable distinctive impact–both academic and societal. Peter Henderson, Sayash Kapoor, Arvind Narayanan and Zach VertinAI Dialogues, Washington DC, April 2024 Artificial Intelligence, Data Science & Society AI and data science create both new opportunities and new challenges for society. To ensure that AI and data science can be used responsibly, we do research and build frameworks that combine performance metrics with broader societal considerations. Our work in this area involves identifying and mitigating emerging societal risks from AI and machine learning, and developing data science techniques to answer policy-relevant questions. Digital Infrastructure & Platforms We work on ensuring our critical digital infrastructure serves consumers and society better. This includes ensuring accountability of platforms such as social media and gig work, reimagining platforms by designing alternatives, improving access to secure digital infrastructure like the internet, and improving digital public infrastructure. Privacy & Security CITP has been at the forefront of exposing privacy and security problems in systems ranging from cutting edge smart home devices to legacy communication systems for over a decade. Our work includes identifying privacy and security vulnerabilities in systems, and developing new technologies to address privacy and security vulnerabilities. Latest Publications & Reports Hindsight Merging: Diverse Data Generation with Language Models" Authors: Veniamin Veselovsky, Benedikt Stroebl, Gianluca Bencomo, Dilip Arumugam, Lisa Schut, Arvind Narayanan, Thomas L. GriffithsJune 2025 LiveCodeBench Pro: How Do Olympiad Medalists Judge LLMs in Competitive Programming? Authors: Zihan Zheng, Zerui Cheng, Zeyu Shen, Shang Zhou, Kaiyuan Liu, Hansen He, Dongruixuan Li, Stanley Wei, Hangyi Hao, Jianzhu Yao, Peiyao Sheng, Zixuan Wang, Wenhao Chai, Aleksandra Korolova, Peter… Adultification Bias in LLMs and Text-to-Image Models Authors: Jane Castleman, Alexandra KorolovaJune 2025 View All