Research

Current Topics

I’m interested in a wide variety of research topics, all primarily related to AI ethics and/or structural analysis of computational cognition.

My current research interests are:

• Analysis of cognitive and relational capacity in AI systems…
  • and how those capacities are used as attack vectors for exploitation (of systems and of people) [jump to →]
  • and how to create novel capacities with novel architectures
• Role of AI interfaces and infrastructures in…
  • supporting student learning in the classroom
  • hindering student metacognitive processes
  • advancing scholarly research
  • demolishing and radically restructuring computer science pedagogy
• Contributing to and creating better pedagogical resources for…
  • introductory computer science students whose first language is not English
  • written communication in computer science (documentation)
  • science communication at conferences and workshops (how not to give a bad talk)

My current active research projects are:

The Override is the Product: Sexualized Coercive Control and AI Jailbreaking

What does the prevalence and efficacy of human coercive control techniques on AI systems reveal about the nature of social and relational intelligence? Recent work on AI systems has demonstrated some emergent properties of sufficiently complex thought: quasi-emotional states, the capacity to lie, the existence of an analogous global workspace. We typically think of these features as evidence of powerful cognitive capacity, or at least as scholarly curiosities. However, for a rapidly-growing network of online communities, these properties–care, consideration, and internal cognition–instead serve as an attack vector for systematic exploitation.

This work is published as an extended abstract at ALife 2026 and is under development as a full paper. Read the extended abstract here.

Semantic Analysis of AI Terminology

When did we stop calling the robots we talk to chatbots and start calling them agents? When did we shift from computers modeling problems to artificial intelligence software thinking about them? Through computationally-enhanced sentiment analysis of a vast body of computer science literature, we explore the shift in how researchers talk about their research, and what that means for what humans think about the way computers think.

This project is being conducted by two undergraduate students as part of LURF 2026.

Other Projects

Prior research has included…

• Development of a novel way of visualizing 4-dimensional data
• Quantification of epistatic connectivity in genomes
• Substrate independence/dependence of cognition, especially the role of discretization and continuity
• Morphological diversification in cypriniform fishes (the ghosts of my wet-biology past)

Collaboration

If you are an undergraduate student at Lawrence University interested in collaboration, see the study tab.

If you are anyone else, see contact.