RESEARCHInvestigateNEXT 12 MONTHS
The Hidden Puppet Master: Predicting Human Belief Change in Manipulative LLM Dialogues
arXiv cs.CL — Computation and Language
Factual evidence
What the source reports
Research introduces PUPPET, a taxonomy and resource to predict human belief change in dialogues with manipulative LLMs.
Open sourceOneBench interpretation
Institutional assessment
So what
Understanding how LLMs can subtly steer user beliefs is critical for G-SIBs building compliant, high-trust AI applications and managing downstream reputational risk.
Do what
This research provides a framework for evaluating LLM outputs for subtle manipulation, which your model risk team should integrate into responsible AI guidelines for user-facing applications.