From Facebook’s data negligence to the first fatalities involving autonomous vehicles, 2018 has seen the public awaken to the profound security implications of widespread AI and large scale data collection. We propose a purposeful convening at Berkeley to reorient cybersecurity research and pedagogy towards protecting civil institutions through robustness, fairness, and systems theory. Berkeley’s Graduates for Engaged and Extended Scholarship around computing and Engineering (GEESE) will lead a research cluster to produce concrete reforms and interventions in R1 graduate curricula. We seek to train the next generation of Ph.D. graduates to tackle the security problems posed by autonomous systems in distinct social spaces. This initiative will furthermore empower Berkeley’s leading disciplinary voices to reimagine the cybersecurity landscape, revealing new opportunities for conceptual integration, specialization, and cross-fertilized growth. As the one-year seed grant concludes, GEESE will produce a strategic plan and policy whitepaper comprising 3-5 policy recommendations from the research cluster, to be shared with affiliates at other R1 institutions, the National Science Foundation, and think tanks focused on improving the cybersecurity talent pipeline. We will also present findings through the creation of an interactive website, connecting cross-national pedagogy innovators by suggesting implementation strategies and collating experiences. The website’s public reception and influence will serve as a metric for success in applying the “GEESE model” of student-led curriculum and conceptual innovation at other institutions.