Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Scoping Review of the Ethical Perspectives on Anthropomorphising Large Language Model-Based Conversational Agents

Andrea Ferrario, Rasita Vinay, Matteo Casserini, Alessandro Facchini

TL;DR

The paper addresses the ethical implications of anthropomorphising LLM-based conversational agents, a phenomenon with both engagement benefits and risk of deception and overreliance. It conducts a scoping review across multiple databases to map how anthropomorphisation is defined, framed, and studied, and to identify methodological approaches linking interaction effects to governance guidance. The authors provide a consolidated conceptual map, categorize recurring risk pathways and normative stakes, and propose a research and governance agenda with actionable design and policy recommendations. The findings highlight a convergence on attribution-based definitions but divergent operationalizations, and reveal a need for longitudinal, governance-ready research that connects cues to measurable outcomes beyond engagement.

Abstract

Anthropomorphisation -- the phenomenon whereby non-human entities are ascribed human-like qualities -- has become increasingly salient with the rise of large language model (LLM)-based conversational agents (CAs). Unlike earlier chatbots, LLM-based CAs routinely generate interactional and linguistic cues, such as first-person self-reference, epistemic and affective expressions that empirical work shows can increase engagement. On the other hand, anthropomorphisation raises ethical concerns, including deception, overreliance, and exploitative relationship framing, while some authors argue that anthropomorphic interaction may support autonomy, well-being, and inclusion. Despite increasing interest in the phenomenon, literature remains fragmented across domains and varies substantially in how it defines, operationalizes, and normatively evaluates anthropomorphisation. This scoping review maps ethically oriented work on anthropomorphising LLM-based CAs across five databases and three preprint repositories. We synthesize (1) conceptual foundations, (2) ethical challenges and opportunities, and (3) methodological approaches. We find convergence on attribution-based definitions but substantial divergence in operationalization, a predominantly risk-forward normative framing, and limited empirical work that links observed interaction effects to actionable governance guidance. We conclude with a research agenda and design/governance recommendations for ethically deploying anthropomorphic cues in LLM-based conversational agents.

A Scoping Review of the Ethical Perspectives on Anthropomorphising Large Language Model-Based Conversational Agents

TL;DR

The paper addresses the ethical implications of anthropomorphising LLM-based conversational agents, a phenomenon with both engagement benefits and risk of deception and overreliance. It conducts a scoping review across multiple databases to map how anthropomorphisation is defined, framed, and studied, and to identify methodological approaches linking interaction effects to governance guidance. The authors provide a consolidated conceptual map, categorize recurring risk pathways and normative stakes, and propose a research and governance agenda with actionable design and policy recommendations. The findings highlight a convergence on attribution-based definitions but divergent operationalizations, and reveal a need for longitudinal, governance-ready research that connects cues to measurable outcomes beyond engagement.

Abstract

Anthropomorphisation -- the phenomenon whereby non-human entities are ascribed human-like qualities -- has become increasingly salient with the rise of large language model (LLM)-based conversational agents (CAs). Unlike earlier chatbots, LLM-based CAs routinely generate interactional and linguistic cues, such as first-person self-reference, epistemic and affective expressions that empirical work shows can increase engagement. On the other hand, anthropomorphisation raises ethical concerns, including deception, overreliance, and exploitative relationship framing, while some authors argue that anthropomorphic interaction may support autonomy, well-being, and inclusion. Despite increasing interest in the phenomenon, literature remains fragmented across domains and varies substantially in how it defines, operationalizes, and normatively evaluates anthropomorphisation. This scoping review maps ethically oriented work on anthropomorphising LLM-based CAs across five databases and three preprint repositories. We synthesize (1) conceptual foundations, (2) ethical challenges and opportunities, and (3) methodological approaches. We find convergence on attribution-based definitions but substantial divergence in operationalization, a predominantly risk-forward normative framing, and limited empirical work that links observed interaction effects to actionable governance guidance. We conclude with a research agenda and design/governance recommendations for ethically deploying anthropomorphic cues in LLM-based conversational agents.
Paper Structure (43 sections, 4 figures, 4 tables)

This paper contains 43 sections, 4 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Working Definitions
  3. Related Work
  4. Methodology
  5. Databases and Sources
  6. Search Strings
  7. Screening and Eligibility
  8. Data Charting and Synthesis
  9. Results
  10. Characteristics of Included Studies
  11. Answering RQ1: Conceptual Foundations
  12. Theoretical foundations
  13. Operationalization
  14. Answering RQ2: Ethical Challenges and Opportunities
  15. Normative stakes
  16. ...and 28 more sections

Figures (4)

  • Figure 1: High-level search string structured into four blocks.
  • Figure 2: PRISMA-ScR flow diagram of this scoping review.
  • Figure 3: Answering RQ1. A scheme summarizing the conceptual foundations of anthropomorphisation from the corpus of included manuscripts.
  • Figure 4: Answering RQ3. Methods and strategies for anthropomorphisation governance across the deployment lifecycle: pre-deployment, during deployment, and post-deployment.