Relational agents
TAICU studies how intelligent systems should behave in relationships. This includes proactive initiative, conversational repair, boundaries, personality, and the accumulation of persistent context.
Our research
Hug
- Status
- Current project
- System
- Relational agent
- Interface
- Messaging
- Public record
- hug.wtf ↗
A persistent AI friend in the places people already talk.
HUG explores whether an AI can be socially present without waiting to be summoned. It operates through familiar messaging interfaces, maintains durable context, and is designed around short, conversational exchanges.
HUG lets TAICU investigate memory, personality, proactive interaction, emotional expectations, and the ways relational systems can fail.
Visit Hug ↗Research agenda
Durable personal context
TAICU studies what an AI should remember, forget, revise, or refuse to retain. The lab seeks persistent context that supports continuity, including memory decay and memory correctability.
Safety in significant interactions
Benchmark-only evaluation does not capture all the risks relational systems create. TAICU studies interaction safety: influence, dependency, emotional failure, and recovery through longitudinal evaluation over time and in context.