NCSU SoS Lablet Quarterly Executive Summary

A. Fundamental Research
High level report of result or partial result that helped move security science forward-- In most cases it should point to a "hard problem". These are the most important research accomplishments of the Lablet in the previous quarter.

We continued to produce science of security outcomes. The following are the major contributions from Lablet projects.

• We implemented and evaluated our hybrid machine learning framework based on 35 real-world attacks on 24 common applications. We showed that our approach produces a false alarm rate that is lower by more than one order of magnitude (e.g., from 13% to 0.6%) compared to current supervised models.
• We enhanced our framework for identifying "UIA-enabling" mobile apps, i.e., apps that enable information access about users and others in a way that violates user expectations. Our framework identified 83 UIA-enabling apps from the given seed datasent and found an additional 48 UIA-enabling apps via snowballing. Through manual examination, we found that nearly two-thirds of these apps are truly UIA-enabling.
• We examined FOSS vulnerability reports and reportters. We found that 20% of reporters produce around 80% of the reports; first-time reporters are active contributorsbut the most productive reporters specialize with respect to the project and vulnerability types.

B. Community Engagement(s)
Research interaction in the community including workshops, seminars, competitions, etc.

We brought up the Science of Security in a variety of fora, including

• Two distinguished lectures we hosted this quarter on Ethics and Safety of AI involved themes relating to science of security; one lecture on regulations and risk and the other on safety envelopes as a way to provide guarantees on behavior in a tractable manner.
• Discussions with non-lablet colleagues locally and at other universities.
• Engagement with a local startup, Airgap Inc., regarding security in convergent Information Technology/Operational Technology networks for secure manufacturing.

C. Educational Advances
Impact to courses or curriculum at your school or elsewhere that indicates an increased training or rigor in security research.