kayleigh bishop’s personal website

about me

I’m an AI researcher specializing in human-AI interaction, LLM evaluation, and interpretable machine learning systems. My research investigates how structured representations, interactive AI systems, and empirical evaluation can improve reasoning, transparency, and decision-making in human-AI collaboration. Until recently, I was in the Ph.D. program at the University of Colorado Boulder (where I got my M.S.), working in the CAIRO group; this year, I decided to depart the program and relocate back to the northeast to pursue applied work in industry.

Before starting grad school, I got my B.S. in Cognitive Science from Yale University. As a first-generation student from a low-income background, I’m personally passionate about addressing educational inequality and making scientific research accessible and inclusive to all.

My preferred pronouns are they/them, but any are fine by me. I publish under the name Kayleigh Bishop, but most people call me Kaleb.

my research

I study how humans and AI systems can work together more effectively, with a focus on evaluating and designing the interactions between them. My forthcoming paper examines task handover and decision-making in collaborative settings, including a project on mental model reconciliation that shows how structured AI support can reduce the cost of staying informed while cutting down on the misinformation that ad hoc AI use tends to introduce. More broadly, I’m interested in the design properties of human-AI interaction — how agency is allocated between people and systems, how transparency is built into an interface, and how a system’s framing shapes trust and understanding. I bring this same lens to questions in education and public communication, where I’m interested in how AI-assisted tools can support metacognition and learning rather than just delivering answers.

contact me

I’m always looking to connect with new people! Please feel free to email me at kaleb.bishop[at]colorado.edu.