Twisters Management Consulting LLC published a paper investigating machine psychology and persona-based dialogue generation, demonstrating that large language models exhibit a systematic bias toward persona overuse.
Public source
Publisher name
Public post
Investigates machine psychology and persona-based dialogue generation (PDG). The authors demonstrate that LLMs exhibit a systematic bias toward "persona overuse"—incorpo…
Company
Twisters Management Consulting LLC
Organizational Development - People Analytics - Investigative Psychology and OSINT
- Industry
- Business Consulting and Services
- Location
- Williston, US
- Company size
- 1 employee
About Twisters Management Consulting LLC
Human Capital Is Our Focus. The name is "Twisters Management Consulting" because "Twisters" is one of our six Appaloosa horses who assist us as co-trainers for Leadership and Personal Impact.
See moreLatest activity
Latest activity from Twisters Management Consulting LLC
7 signals
Research & Knowledge
Twisters Management Consulting LLC published an empirical study investigating when persona prompting succeeds versus fails in aligning model responses with diverse human sub-populations, evaluating across four national social surveys and six LLMs.
Research & Knowledge
Twisters Management Consulting LLC published a research paper titled From Deceptive Outputs to Deceptive Mechanisms: A Causal Framework for Language-Model Deception Research.
Research & Knowledge
Twisters Management Consulting LLC published a research paper titled Moral Competence Before Moral Content: Why LLM Agents Lack the Prerequisites for Coherent Alignment evaluating nine frontier models across four structural criteria.
Discover more
Similar signals
Similar public activity from other companies.
Research & Knowledge
NEC Laboratories America, Inc.
NEC Laboratories America, Inc. published a paper titled Mind the Gap in Cultural Alignment: Task-Aware Culture Management for Large Language Models on July 7, 2026.
Research & Knowledge
Munich Center for Machine Learning
Munich Center for Machine Learning published a new preprint analyzing 89,253 outputs from 12 LLMs in English, French, and Chinese across 18 occupations and three task conditions.
Research & Knowledge
LBZ Advisory LLC
LBZ Advisory LLC published The Bias Advantage, a paper discussing how AI can inherit bias as operating logic.
Research & Knowledge
Association for Contextual Behavioral Science
Association for Contextual Behavioral Science published a new study finding that LLMs like GPT 5.2 and Grok 4.1 show near-complete contingency insensitivity, similar to human rule-governed behavior.
Research & Knowledge
Fraunhofer FIT Generative AI Lab