This project focuses on developing the next generation of network scanning tools and methodologies for more efficiently finding all Internet devices and software, safely uncovering vulnerabilities and misconfigurations in them, and identifying device owners so that they can be notified before they are attacked. This project will introduce fundamentally new methodologies in three areas. First, drawing on advances in artificial intelligence and software fuzzing, the project will build new techniques for finding network scan probes that safely identify hardware and software manufacturers, products, and fine-grained versions. The project will build agentic approaches to scalably uncover human misconfigurations that could not otherwise be programmatically identified. Second, new methods to uncover relationships between Internet entities and to extract device owners will be developed. Third, building on the context and relationships derived about Internet assets, new predictive methods for finding Internet services as they come online will be developed. This project will provide new techniques for networking and security researchers to better understand the Internet, for U.S. organizations to more quickly protect themselves against attacks. The project will develop new curriculum to prepare computer science students to work in cybersecurity by providing them hands-on opportunities to understand Internet-security in practice as well as provide research opportunities to both undergraduate and graduate students. Publications, open source software, and other research outputs will be made publicly accessible at http://esrg.stanford.edu/. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.