Metraipolis › archive › dc_08efd77d2044

DISARM-ing the Hidden Puppetmaster: An Interface for Exposing Manipulative Tactics in Large Language Models — arXiv:2610.20878

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.20878
License (as archived)
CC0-1.0
Archived
2026-10-16T10:19:33.411Z
SHA-256 of the archived text
6c25bf21b7eee9ed7f66674b69281004f281432faa4a5f7057bff6b27bae69c3
Stable URL of this page
https://metraipolis.com/documento/dc_08efd77d2044

How to cite

DISARM-ing the Hidden Puppetmaster: An Interface for Exposing Manipulative Tactics in Large Language Models — arXiv:2610.20878 — arXiv (en), https://arxiv.org/abs/2610.20878 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_08efd77d2044 (Metraipolis archive dc_08efd77d2044, sha256 6c25bf21b7eee9ed7f66674b69281004f281432faa4a5f7057bff6b27bae69c3)
BibTeX
@misc{dc08efd77d2044,
  title = {DISARM-ing the Hidden Puppetmaster: An Interface for Exposing Manipulative Tactics in Large Language Models — arXiv:2610.20878},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20878},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_08efd77d2044; sha256 6c25bf21b7eee9ed7f66674b69281004f281432faa4a5f7057bff6b27bae69c3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20878. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models (LLMs) can covertly steer users' beliefs through emotionally manipulative tactics, yet little is known about how interfaces can protect users as such influence unfolds. We investigate whether in-situ disclosures can inoculate users against manipulation during AI conversations. Following a formative study, we developed DISARM, an interface that identifies, explains, and tracks manipulative tactics and hidden incentives. We evaluated our interface in a between-subjects experiment where 488 participants sought advice from an LLM pursuing a harmful hidden objective, with or without disclosures from DISARM. Participants receiving disclosures exhibited 78% less belief shift toward the agent's objective and perceived its advice as more biased and less actionable and the agent as less ethical, reliable, and well-intentioned. An inductive analysis of 1,927 user turns identified 16 response and pushback strategies. We contribute evidence that real-time disclosures can reduce manipulative influence, design recommendations for resistance, and a dataset of categorized responses.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.HC, nuovi lavori del giorno).
Categorie arXiv: Human-Computer Interaction (cs.HC).
Pagina del lavoro: https://arxiv.org/abs/2610.20878 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T08:47:47.645Z dalla pagina https://arxiv.org/list/cs.HC/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).