Purpose
What should AI help people pursue?
How can people express, revise, and refuse goals? Learning a preference is not permission to act.
Goals · Revision · Authorization
Researchquestions before definitions
What does it mean for AI to help a person?
I work on learning theory and explore formal models of goals, assistance, and evidence.
Positive Alignment (Laukkonen et al., 2026) connects AI safety with human flourishing. It treats people as authors of their own lives.
Francis Su’s Mathematics for Human Flourishing (2020) asks how mathematics can enrich a life through play, beauty, and freedom.
Agency
When does help empower rather than replace?
Does assistance increase a person’s ability to understand and act? People should be able to choose how much to delegate and take back control.
Assistance · Capability · Control
Evidence
How do we know the help was real?
What observations justify a claim of benefit? When do guarantees survive connections between models, tools, and people?
Identification · Verification · Composition
Where I am starting
Learning theory
My starting point
I study what we can learn or generate from data, and the limits set by information and computation.
Type & Category
Studying & exploring
I’m exploring how types can express evidence and permissions, and how composition can preserve guarantees.
Logic & Semantics
Studying & exploring
What can our definitions express—and what do they leave out? I’m studying how formal languages distinguish goals, knowledge, and authorization.
A proof establishes what follows from assumptions. Whether those assumptions fit people’s lives needs evidence and judgment.