Characterizing the Extent of Harm Using Real User Data

Overview

This track analyzes real user chatbot conversations to characterize how harm emerges and how often it appears in practice. The goal is to ground risk analysis in concrete, de-identified examples.

Figure 3 from the Characterizing Delusional Spirals study.
Figure 3. Characterizing study overview.

TODO: add more

Citation

@article{moore_characterizing_2026,
  title = {Characterizing Delusional Spirals through Human-LLM Chat Logs},
  author = {Moore, Jared and Mehta, Ashish and Agnew, William and Anthis, Jacy Reese and Louie, Ryan and Mai, Yifan and Yin, Peggy and Cheng, Myra and Paech, Samuel J. and Klyman, Kevin and Chancellor, Stevie and Lin, Eric and Haber, Nick and Ong, Desmond},
  booktitle = {Working draft},
  note = {Draft citation placeholder.},
}