A4 Lab
actions against advanced adversaries
The A4 Lab (actions against advanced adversaries) at University of Arizona is directed by Professor Zhuolin Yang. We research at the intersection of Security, Usability, and Artificial Intelligence (AI). In a world of advanced adversaries (now armed with powerful AI models and computing systems), we aim to identify emerging threats and build usable tools that help individuals maintain agency over their security and privacy.
📣 We are looking for students interested in Security and Usability research. If you're interested, please email Professor Yang (zhuoliny at arizona dot edu) and mention Prospective PhD, Master or Undergraduate Student in the subject line.
Current Research Interests
AI-enhanced Privacy Attacks
Training AI typically demands large quantities of labeled data, which is costly and often infeasible in adversarial settings. We identify techniques attackers might use to minimize the costs and exploit AI to launch practical enhanced attacks.
Past Project:
Wearable Privacy Protections
With ubiquitous sensors and powerful AI-driven interpretation, personal privacy is increasingly susceptible to adversarial inference. We study how an individual's personal computing environment (e.g., wearables) can be used to counteract privacy inferences in everyday life.
Past Project:
Security in XR Systems
Traditional tools (e.g., video calls) lack key nonverbal cues like body language, limiting the depth of digital communication. XR systems address this by recreating immersive, lifelike experiences. Yet, this realism also introduces new risks, as misuse of XR interfaces can enable impactful attacks.
Past Project:
- A Keystroke Inference Attack in Virtual Reality Systems
- Immersive Hijacking in Virtual Reality Systems