Skip to main content

User Narrative Study for Dealing with Deceptive Chatbot Scams Aiming to Online Fraud

Veisi, Omid and Kazemian, Khoshnaz and Gerami, Farzaneh and Mirzaee Kharghani, Mahya and Amirkhani, Sima and Du, Delong K. and Stevens, Gunnar and Boden, Alexander (2025) — Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems

Citation activity

Total citations · Google Scholar
Unavailable
Average per calendar year
Citation count unavailable

Counts reflect Google Scholar’s coverage and do not measure research quality. Source, calculation and limitations

View Original Source (opens in a new tab)

DOI: 10.1145/3706599.3720152 (opens in a new tab)

Tags:   Chatbot Impersonation, Chatbot Scams, LLMs, Narrative Interviews, Online Fraud, Romance Scams, Scam Detection, Social Media Scams
What do these research terms mean?
Preprint
A manuscript shared before formal peer review and publication. Check whether a later published version is available.
Dataset
A collection of data or examples for others to inspect or reuse. It can appear in Library search and topic mapping, but RSRC does not use dataset records as evidence in Research Insights.
Dissertation or thesis
Research submitted for an academic degree. This describes its format, not its reliability.
Journal article
An article published in a journal. This label alone does not establish peer review, study quality, or how well the findings apply elsewhere.
Qualitative research
Examines experiences, meanings, or processes, often through interviews or observations. It can explain how something happens without estimating how common it is.
Quantitative research
Uses numerical measurements to describe patterns or test relationships. A relationship between two measurements does not by itself show that one causes the other.
Systematic review
Uses a planned, documented method to find and assess research addressing a question. Its conclusions still depend on the included studies and what the search covered.
Meta-analysis
Statistically combines results from multiple studies. Combining studies does not remove weaknesses in their design or make unlike populations interchangeable.
Not classified
This record has no recognized label in this filter. It does not mean the publication used no method, or that no research exists.

Definitions draw on DataCite resource types; Cochrane review methods; NLM: association and causation. RSRC’s dataset and classification rules are explained in our methodology.

Citation tools
Download RIS Download BibTeX 


              
              

            

          

        


      

      

    

        
Transparency

        
Evidence and review status

        
This page contains AI-generated content. No human content review or subject-matter-expert review is recorded.

    


    

                    

                
Source basis

                

                    Downloaded PDF
                

            

                    

                
Source updates

                

                    No notice found at last check
                

            

                    

                
AI-generated page content

                

                    Yes
                

            

                    

                
Automated checks

                

                    Passed
                

            

                    

                
Administrative approval

                

                    Yes
                

            

                    

                
Human content review

                

                    Not recorded
                

            

                    

                
Subject-matter-expert review

                

                    Not recorded
                

            

            


    

        
            How this was prepared
        
        

                            

                    
Source basis

                    

                    
RSRC downloaded and privately stored a copy of the paper for internal analysis. The PDF is not offered to viewers from this page.


                                            
    
                                                            
  • 
                                    PDF added to RSRC:
                                    
                                
    • 
                                                    

                    
                                        

                

                            

                    
Source updates

                    

                    
No incoming update notice was found in the dated Crossref response. Coverage is incomplete, particularly for corrections and expressions of concern; this is not a guarantee that the source is valid or unchanged.


                                            
    
                                                            
  • 
                                    Last source-status attempt:
                                    
                                
    • 
                                                    

                    
                                        

                

                            

                    
AI-generated page content

                    

                    
AI-generated research notes displayed on this page: Synopsis, Identified gaps, Methods, Limitations, Future work. The paper itself is not described as AI-generated.


                                            
    
                                                            
  • 
                                    Document analysis recorded:
                                    
                                
    • 
                                                            
  • 
                                    Synopsis generation recorded:
                                    
                                
    • 
                                                            
  • 
                                    Page record updated:
                                    
                                
    • 
                                                    

                    
                                        

                

                            

                    
Automated checks

                    

                    
The current, source-bound synopsis passed the recorded versioned publication checks.


                                            
    
                                                            
  • 
                                    Checks completed:
                                    
                                
    • 
                                                    

                    
                                            

                            
                                View passed checks (3)
                            
                            
    
                                                                    
  • Length, completeness, repetition, refusal, boilerplate, and active-markup screening
    • 
                                                                    
  • Numerical claims checked against the available source text
    • 
                                                                    
  • English-source lexical grounding check
    • 
                                                            

                        

                                        

                

                            

                    
Administrative approval

                    

                    
An authenticated administrator approved the bibliographic record for public Library display. This is not a review of every research claim.


                                            
    
                                                            
  • 
                                    Approved for public display:
                                    
                                
    • 
                                                    

                    
                                        

                

                            

                    
Human content review

                    

                    
No human review is recorded for the AI-generated content displayed on this page.


                    
                                        

                

                            

                    
Subject-matter-expert review

                    

                    
RSRC has not recorded review of this content by a subject-matter or methods expert.


                    
                                        

                

                    

    


    

        Review-state definitions

        Found a possible error?
        Request a correction.
    



    



              

        
                  

            
Synopsis

            
This paper reports a qualitative study exploring how users perceive and identify deceptive chatbots involved in online scams. The authors aim to understand the user perspective on chatbot impersonation, focusing on how such bots manipulate interactions to mislead people and what cues people rely on to distinguish bots from authentic human actors. The study gathers narratives from 16 individuals who believed they were victims of scam bots, using semi-structured interviews to elicited detailed accounts of contacts, deception tactics, and aftermath. Methodologically, the researchers conducted narrative thematic analysis of interview data. Participants varied in age (24–35) and professional background, and encounters spanned several platforms (Telegram, Instagram, websites, Gmail, and others). Interviews explored the arc of the fraud—initial contact, realization of deception, and post-incident reflections—along with perceived patterns that signal non-human interaction. The analysis emphasized temporal and communicative cues, such as consistent timing, immediate or formulaic responses, single-sentence messages, repetitive messaging, and a lack of genuine emotional depth. Principal findings indicate that chatbots impersonate a range of social roles (job recruiters, wealthy seniors, finance traders, famous celebrities, acquaintances) to provoke trust, financial decisions, or emotional engagement. Users reported recognizing bots through mechanical language, predictable responses, and odd patterns such as synchronous activity or abnormal promptness, especially when topics diverged from what a human would normally handle. The study discusses implications for cybersecurity and UX design, suggesting that awareness of these cues can support safer digital interactions. It notes limitations, including focus on victim narratives and the evolving nature of AI, and points to future work on expert perspectives, demographic susceptibility, and the potential role of large language models in scams. The authors propose a framework for improving scam detection informed by user experiences.

          

        
                  

            
Identified Gaps

            
The paper identifies an underexplored human perspective on chatbot deception: how users perceive and detect it. It also notes a lack of observed examples of LLM programs targeting individuals, likely because such systems are relatively new. Further evidence is needed on demographic and cultural differences in susceptibility and targeting by advanced AI-enabled scammers.

          

        
                  

            
Methods

            
The study used semi-structured narrative interviews with 16 adults who believed they had experienced chatbot scams on social-media platforms, websites, or email. Interviews covered the fraud narrative, challenges before, during, and after the incident, emotional and behavioral reactions, detection, behavior changes, and prevention advice. Researchers conducted narrative thematic analysis, initially analyzing four accounts individually and then using collaborative co-analysis to identify cross-case narrative patterns.

          

        
                  

            
Limitations

            
The study focuses only on victims' accounts and what happened during fraud, excluding interviews with experts or scammers. Its preliminary qualitative sample comprised 16 self-identified chatbot-scam victims aged 24–35, so findings reflect subjective recollections and may not generalize broadly. The authors also found no examples of LLM programs used to target individuals, limiting conclusions about newer LLM-enabled scams.

          

        
                  

            
Future Work

            
Interview experts and scammers about chatbot fraud before, during, and after incidents. Investigate whether and how LLM-based chatbots are used in scams and develop responses to such use. Examine how gender, age, socioeconomic status, and cultural background affect susceptibility and targeting. Apply narrative coding to expert narratives to identify qualitative themes and psychological and social dynamics.

          

              

    
    

        
Related Research Insights

        

            See how this publication connects to RSRC's living evidence syntheses through current citations and research-topic mapping.