Lauren Project
Preliminary LLM research, 2026. By Rivian Pratama. A first exploration into how LLMs can understand humans.
The question behind this project is simple and it is whether a machine can really understand a person. I believe the answer starts with the words we already share with the people closest to us.
The system gives raw chat transcripts to the model as context. The advice comes from the actual conversation history. The responses follow the emotional register of the messages, which made the warmth uncanny and the work controversial because it trades privacy for empathy.
I wanted to see if a machine could learn what makes a relationship work just from the words two people share.