DEMO
Yiliu Tang, Mengke Wu, Jason Situ, Jingyi Liu, Andrea Yaoyun Cui, Yun Huang
The integration of generative AI with Extended Reality (XR) technologies has unlocked unprecedented capabilities, giving users enhanced cognitive, sensory, and environmental control—effectively conferring “superpowers” in immersive digital spaces. Our CHI 2026 literature review paper examines both the benefits and potential risks of these LLM‑enabled XR superpowers. It makes two main contributions: (i) a synthesized taxonomy of these superpowers and their associated risks, and (ii) a set of design guidelines and a forward research agenda derived from that synthesis. Through a multi‑phase analysis of 135 recent works in the field, we categorize superpowers into internal (cognitive and sensory enhancements) and external (environmental and social manipulations) and illustrate how they amplify human abilities in domains such as healthcare, education, and professional training. We also analyze the risks specific to each superpower, highlighting vulnerabilities in user autonomy, data security, and ethical transparency. This work aims to guide stakeholders in harnessing the potential of XR while mitigating the socio‑technical risks of this emerging landscape.
In addition to the poster, we will demonstrate LLM‑powered smart‑glasses applications that support users in live‑streaming spatial knowledge while explicitly embedding ethical safeguards—such as privacy‑preserving capture, informed consent mechanisms, and responsible data handling.