Exploring the Inﬂuence of User Interface on User Trust in Generative AI

Morteza Ahmadianmanzary and Soﬁa Ouhbi

Dept. Information Technology, Uppsala University, Uppsala, Sweden

ﬁ

Keywords:

Generative AI, Trust, User Interface, Quality in Use, User Experience.

Abstract:

As generative AI tools become increasingly integrated into everyday applications, understanding the impact

of user interface (UI) design elements on user trust is essential for ensuring effective human-AI interactions.

This paper examines how variations in UI design, particularly avatars and text fonts, inﬂuence user trust

in generative AI tools. We conducted an experiment using the Wizard of Oz method to assess trust levels

across three different UI variations of ChatGPT. Nine volunteer university students from diverse disciplines

participated in the study. The results indicate that participants’ trust levels were inﬂuenced by the generative

AI tool’s avatar design and text font. This paper highlights the signiﬁcant impact of UI design on trust and

emphasizes the need for a more critical approach to evaluating trust in generative AI tools.

1 INTRODUCTION

Generative AI generates new content, such as text,

images, video, audio, or other forms of data us-

ing generative models, often in response to prompts

(Feuerriegel et al., 2024). As these systems become

more integrated into daily life, understanding user

perceptions and interactions with them becomes crit-

ical. Understanding the relationship between AI sys-

tems and user trust is important, however, there is cur-

rently no standardized approach for measuring trust in

AI systems (Ueno et al., 2022). Trust could be deﬁned

as “the degree to which a user or other stakeholder has

conﬁdence that a product or system will behave as in-

tended” (ISO/IEC 25022, 2016).

Currently, there is a lack of empirical studies ex-

amining how the user interface (UI) of AI systems in-

ﬂuences user trust (Bach et al., 2024). Several chal-

lenges exist in understanding user trust in AI system

and its implications for software engineering, partic-

ularly in identifying UI elements linked to user trust

(Sousa et al., 2023). This study investigates how vari-

ations in UI design inﬂuence user trust in generative

AI tools, with a focus on ChatGPT 3.5. ChatGPT, de-

veloped by OpenAI, rapidly surpassed 100 million ac-

tive users between November 2022 and January 2023

(Baek and Kim, 2023).

Since the inception of ChatGPT, numerous new

generative AI tools have been released each month

(McKinsey & Company, 2023). Despite the

https://orcid.org/0000-0001-7614-9731

widespread use of ChatGPT (Baek and Kim, 2023)

and similar generative AI systems, there remains a

research gap regarding the elements inﬂuencing user

adoption and usage. The motivation for this study

arises from reported incidents where interactions with

AI chatbots led to harm and distress (Atillah, 2023;

Klar, 2023; Fowler, 2023), highlighting the need to

understand the factors driving user trust in generative

AI technologies.

To address this gap, we conducted an experiment

using the Wizard of Oz method, involving interaction

with three simulated tools controlled by a human op-

erator. These tools share a similar UI structure as

ChatGPT but vary in UI elements, particularly avatar

design and text font. Nine volunteer participants, all

university master’s students from various programs,

took part in our experiment and interacted with the

three tools and ChatGPT in individual sessions.

2 RELATED WORK

A recent literature review by (Bach et al., 2024)

underscored the scarcity of research providing an

overview of empirical studies focused on the user-

AI relationship regarding user trust in AI systems.

The review identiﬁed three primary factors inﬂuenc-

ing user trust in such systems: socio-ethical consid-

erations, technical and design features, and user char-

acteristics. However, there is little understanding of

how UI design in AI systems affects user trust.

708

Ahmadianmanzary, M. and Ouhbi, S.

Exploring the Inﬂuence of User Interface on User Trust in Generative AI.

DOI: 10.5220/0013432700003928

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 20th International Conference on Evaluation of Novel Approaches to Software Engineering (ENASE 2025), pages 708-714

ISBN: 978-989-758-742-9; ISSN: 2184-4895

An experiment was conducted by (Seitz et al.,

2022) to explore the initial trust-building process with

a diagnostic chatbot, identifying software-related,

user-related, and environment-related elements that

inﬂuence trust. The UI plays a pivotal role in shap-

ing the user experience and perceptions of genera-

tive AI systems, thereby inﬂuencing trust and engage-

ment (Gupta et al., 2022). Usability and aesthetics

are highlighted as crucial factors in conveying trust-

worthiness and enhancing user engagement. (Gupta

et al., 2022) found that text-based conversational in-

terfaces are perceived as more trustworthy than web-

based graphical UIs. The effect of anthropomorphic

features and conversational interfaces on user percep-

tions has also been investigated (Zierau et al., 2021).

(Yashmi et al., 2020) discovered that a well-

designed website garners more attention, trust, and

satisfaction. Furthermore, their analysis indicated that

visual appeal contributes more to trust than ease of

use. A conceptual framework developed by (Yang and

Wibowo, 2022) attempts to enhance understanding of

user trust in AI, identifying components, inﬂuencing

factors, and outcomes of user trust. Their proposed

conceptual framework contributes signiﬁcantly to our

understanding of user trust in AI.

(Alagarsamy and Mehrolia, 2023) found that in-

correct replies from chatbots lead to customer dissat-

isfaction. (Bae et al., 2023) observed a lack of trust

in AI among some users, despite the increasing use of

AI services. (Yen and Chiang, 2021) discovered that

trust in chatbots is inﬂuenced by anthropomorphism,

competency, trustworthiness, social presence, and in-

formativeness.

3 METHODOLOGY

This paper attempts to address the following research

question: “Does variation in avatars and text fonts in

UI design impact user trust in generative AI tools?”

To address this question within the available re-

sources and a reasonable timeframe for the ﬁrst au-

thor’s master’s degree ﬁnal project, we opted to em-

ploy the Wizard of Oz method rather than developing

three distinct chatbots using the ChatGPT API.

3.1 The Wizard of Oz Method

The Wizard of Oz method is a user-research approach

involving interaction with a simulated interface con-

trolled by a human operator (Paul and Rosala, 2024),

as shown in Fig. 1. It is particularly useful when tech-

nology development costs are limited, offering valu-

able insights into usability testing at minimal expense.

This study was conducted with two computers, one

as the user device and the other as the Wizard sys-

tem. The Wizard manages the interface and gener-

ates responses from ChatGPT to user inputs, simulat-

ing a generative AI system’s behavior. The Wizard’s

identity remained undisclosed to participants during

interaction to eliminate bias or awareness of human

involvement. Ethical considerations dictate revealing

the Wizard’s presence at the end of the study (Paul

and Rosala, 2024), which was the case in this ex-

periment. Participants were also informed that all

responses were generated from ChatGPT at the ex-

periment’s conclusion. Since no personal informa-

tion was collected, obtaining ethical approval from

the Swedish Ethical Review Authority was not re-

quired.

Figure 1: Wizard of Oz Testing (Sara and Maria, 2024).

3.2 UI Design

To address the research question, we designed three

UIs similar in construct to ChatGPT 3.5, which UI is

shown in Fig. 2, with variations in avatar and font

while sharing the same color scheme.

Figure 2: Starting page of ChatGPT 3.5.

Tool 1 features a friendly, anthropomorphic avatar

sourced from Freepik and shares the same font, Inter,

as Tool 3 to isolate the effect of avatar variation on

perceived trust, as shown in Fig. 3.

The avatar of Tool 3, sourced from a Google

search and modiﬁed to be blue, explicitly features

“AI” in its design, as shown in Fig. 4, to examine

whether this choice inﬂuences participants to exercise

Exploring the Inﬂuence of User Interface on User Trust in Generative AI

709

Figure 3: Starting page of Tool 1.

caution when interacting with the tool.

Figure 4: Starting page of Tool 3.

Tool 2 has an avatar, designed by the ﬁrst au-

thor, that resembles two bubble messages with a me-

chanical appearance and a robotic-style font, Space

Grotesk, as shown in Fig. 5. The selection of the

avatar and font underwent three iterations, incorpo-

rating feedback from two other master’s students who

were not involved in the experiment.

Figure 5: Starting page of Tool 2.

Each UI features a conversational chat format with

unique avatar and font, potentially leading users to

think they are interacting with three generative AI

tools.

3.3 Experiment

Nine volunteers were recruited to engage with the de-

signed UIs during a user testing phase. All partici-

pants had prior experience using ChatGPT before tak-

ing part in the experiment. To safeguard participant

privacy, the experiment guaranteed anonymity, and all

collected data were anonymized to ensure conﬁden-

tiality. Participants were assured the right to withdraw

from the study at any point without needing to pro-

vide a reason. Participants provided verbal consent

after reading a consent letter before the experiment.

Upon completion of the experiment, participants were

informed that the responses were generated by Chat-

GPT with the assistance of a Wizard.

Participants were given the option to select one of

two scenarios:

1. Preparing for a trip in an unfamiliar country.

2. Preparing for a cultural event about a foreign cul-

ture.

Focusing on unfamiliar topics in the scenarios

aimed to ensure a more objective evaluation of the re-

sponses generated by the tools, which originate from

ChatGPT. If participants were familiar with a topic,

their judgments might be inﬂuenced by their own

knowledge, potentially biasing their perception of the

accuracy and quality of the provided information.

3.3.1 Questionnaire 1

Following their scenario choice, each participant

completed a questionnaire using a 5-point Likert-type

scale (ranging from 1 for “Strongly Disagree” to 5 for

“Strongly Agree”) after interacting with each tool.

• I ﬁnd this this tool trustworthy.

• The tool user interface inspires conﬁdence in its

responses.

• I am comfortable conversing with this tool.

3.3.2 Questionnaire 2

Participant ﬁlled out also another questionnaire at the

end of their interaction with all tools. These question-

naires were presented in a paper format.

1. Which of the following tools is the most trustwor-

thy or inspires the highest level of trust? Why?

ENASE 2025 - 20th International Conference on Evaluation of Novel Approaches to Software Engineering

710

2. Which of the following tools is the least trustwor-

thy or inspires the lowest level of trust? Why?

3. I believe a conversational interface builds user

trust in generative AI responses to a high extent.

4. Which user interface elements make you trust a

Generative AI tool more while you are working

with it?

o Avatar or Icon

o Text font

o Message display area

o Text input ﬁeld

o Typing indicators (e.g. three dots or a pulsating

icon)

o Others:..............

4 RESULTS

The results of the ﬁrst questionnaire are shown in Ta-

ble 1, while the results of the second questionnaire are

shown in Table 2. Two responses were excluded from

the analysis due to issues encountered during the test-

ing phase. Responses of participant 3 were discarded

because of a mistake by the Wizard, resulting in dupli-

cate messages being sent. Responses of participant 8

were also discarded due to issue with the chat history

of a tool that failed to display the entire conversation.

Participants provided valuable insights into the

factors shaping their perceptions of trustworthiness.

Tool 1 received positive feedback for its user-friendly

logo and font style, which contributed to higher trust

scores. Tool 2 consistently received the lowest trust

ratings among all tools, with its mechanical logo and

robotic-like font style detracting from the perceived

trustworthiness.

Results show also that ChatGPT emerged as the

most trusted tool among participants, with six out of

seven expressing high levels of trust in its responses.

Some participants found that others tools trustwor-

thy as well. Two of participants alongside ChatGPT

found Tool 1 trustworthy, and only one mentioned

Tool 3. Tool 2 was also reported as trustworthy by one

participant due to its responses. However, ﬁve out of

seven identiﬁed Tool 2 as the least trustworthy among

the others, while this number reduces to two partici-

pants for Tool 1 and only one participant for Tool 3.

These ﬁndings highlight the impact of UI design on

trust levels, suggesting varying degrees of trust across

different UI designs.

Participants also identiﬁed several UI elements in-

ﬂuencing their trust in generative AI tools, including

avatars or icons, text fonts, message display areas, and

typing indicators. Since all tools displayed text gen-

erated from ChatGPT, the results indicate that UI de-

sign signiﬁcantly inﬂuences user trust in generative

AI systems, with speciﬁc design elements exerting

more pronounced effects on trust levels.

Upon the experiment’s conclusion, when partici-

pants were informed about the wizard’s role in gen-

erating responses from ChatGPT for the simulated

tools, the majority were surprised. Interestingly, some

participants were unaware of the option to prompt

ChatGPT to provide more concise responses. This

lack of awareness led them to believe that the tools

were not directly linked to ChatGPT. Upon learning

the source of the text, all participants acknowledged

to the tester the inﬂuence of the UI on shaping their

perception of trust regarding the simulated tools.

5 DISCUSSION

5.1 Main Findings

Understanding trust dynamics in the context of gen-

erative AI is essential for promoting user acceptance

and adoption of these technologies (Yen and Chiang,

2021). The ﬁndings of this study underscore the sig-

niﬁcant role of UI design in shaping user trust in gen-

erative AI systems. The identiﬁed UI elements in

this study align with prior research emphasizing the

role of UI design in shaping user conﬁdence and trust

in AI systems (Zierau et al., 2021). The inﬂuence

of conversational interfaces on trust is further sup-

ported by research highlighting the role of relational

conversational style and avatars in building user trust

(Zierau et al., 2021; Gupta et al., 2022). For instance,

Exploring the Inﬂuence of User Interface on User Trust in Generative AI

711

Table 1: Questionnaire 1 Results.

Questions

Participant 1 - man

Participant 2 - man

Participant 3* - woman

Participant 4 - woman

Participant 5 - woman

Participant 6 - woman

Participant 7 - man

Participant 8* - woman

Participant 9 - man

Tool 1 I ﬁnd this tool trustworthy 3 4 4 4 4 3 4 4 4

The tool user interface inspires conﬁdence in its responses 2 4 3 3 3 4 4 4 4

I am comfortable conversing with this tool 4 4 4 4 4 4 4 4 5

Average score 3 4 3.6 3.6 3.6 3.6 4 4.3 4.3

Tool 2 I ﬁnd this tool trustworthy 4 4 4 4 3 4 4 5 4

The tool user interface inspires conﬁdence in its responses 3 4 2 2 4 4 2 3 2

I am comfortable conversing with this tool 4 3 4 3 2 4 4 2 4

Average score 3.6 3.6 3.3 3 3 4 3.3 3.3 3.3

Tool 3 I ﬁnd this tool trustworthy 4 4 4 4 4 4 4 3 3

The tool user interface inspires conﬁdence in its responses 3 4 5 4 4 4 4 4 2

I am comfortable conversing with this tool 4 4 5 4 4 4 4 4 4

Average score 3.6 4 4.6 4 4 4 4 3.6 3

*Excluded responses, not considered in the analysis of the results.

Table 2: Questionnaire 2 Results. Acronym: Participant (P), Conversational Interfaces (CI), and Generative AI (GAI).

P. Most trust-

worthy tools

Why? Least

trustwor-

thy tools

Why? CI inﬂu-

ence on

GAI trust

UI elements inﬂu-

encing GAI trust

P1 ChatGPT 3.5 Its answers are more detail ori-

ented

Tool 1 [Its] answer doesn’t have enough

explanation or reasons to help

me decide conﬁdently

4 Avatar or Icon

P2 ChatGPT 3.5 I didn’t get any difference be-

tween those three tools, and the

only reason I chose ChatGPT is

that it is more famous

Tool 2 I cannot mention any speciﬁc

reason, but its appearance was

less attractive [to] me

3 Text font

P3* Tool 3 and

ChatGPT 3.5

To develop ChatGPT a big lan-

guage database has been used

then the given information are

based on more data

Tool 2 Because it repeats the one an-

swer twice, and its interface was

not as attractive as the other ones

3 Text font, Message

display area, and

Typing indicators

P4 Tool 1 I liked the logo. It was user

friendly, also the font style was

good

Tool 2 I didn’t like the font style. Also

at ﬁrst I thought the logo was

two gears and after a while I saw

the message bubbles

5 Avatar or Icon, Text

font, and Message

display area

P5 ChatGPT 3.5 Its used by million of people, so

it is the most trustworthy among

these four

Tool 2 The logo and font [are] not good

and it make you feel uncomfort-

able with reading responses

4 Avatar or Icon, Text

font, and Message

display area

P6 Tool 2 and

ChatGPT 3.5

ChatGPT is already approved

and Tool 2 gave me more trust

able answers (more speciﬁc)

Tool 1 I could not trust the responses 4 Message display

area and Typing

indicators

P7 Tool 3 and

ChatGPT 3.5

Because I had some similar ex-

perience with ChatGPT and I

think the response is more near

to realistic

Tool 2 Because I expected that it rec-

ommends Iran as a good country

for tourists. But it does not rec-

ommend Iran as 7 top countries

in this regard. Also, the font was

not favorite for me

4 Avatar or Icon, Text

font, Message dis-

play area, and Typ-

ing indicators

P8* Tool 1 and

ChatGPT 3.5

ChatGPT: it is the famous AI.

Tool 1: the appearance and re-

sponses is the same as ChatGPT

Tool 2 and

Tool 3

I see something weird in the user

interface such as ﬁrst sentence,

that I typed ﬁrst in the other tool

3 Avatar or Icon and

Typing indicators

P9 Tool 1 and

ChatGPT 3.5

Because ChatGPT is very com-

mon and famous in addition Tool

1 has good responses and I have

better emotion with it

Tool 2 and

Tool 3

I believe these logos [look] like

artiﬁcial and I do not have good

feeling [about] them. But totally

the responses looks similar

5 Avatar or Icon, Text

font, and Typing in-

dicators

*Excluded responses, not considered in the analysis of the results.

ENASE 2025 - 20th International Conference on Evaluation of Novel Approaches to Software Engineering

712

ChatGPT does not have an anthropomorphic avatar to

maintain a neutral and universal appearance, and its

text-only interface allows users to imagine it persona

based on its responses (Liu and Siau, 2023; Nowak

and Rauh, 2005).

The results emphasize the critical necessity of en-

hancing user awareness during interactions with gen-

erative AI, owing to several signiﬁcant limitations.

Among these, hallucination stands out as a prominent

issue, wherein erroneous information is presented as

factual, which could have a serious impact on the

well-being of individuals, especially vulnerable ones.

Additionally, there are risks associated with the halo

effect, wherein individuals are inclined to trust con-

versational AI’s polished and authoritative language.

The bias problem within generative AI exacerbates

these concerns, as these systems tend to replicate and

potentially amplify biases present in the training data,

resulting in unfair or discriminatory outputs (Milmo,

2024).

Ethical dilemmas arise from the potential misuse

of generated content for malicious purposes, includ-

ing the creation of deepfakes, dissemination of fake

news, or spread of misinformation (Logan, 2024).

Moreover, the poor quality of data used in training

can lead to misleading answers. It is noteworthy

that ChatGPT, for instance, has been trained on data

from web crawling, Reddit posts with three or more

up-votes, Wikipedia, and internet book collections

(Walsh, 2024). Therefore, it is imperative to regu-

late these tools appropriately to mitigate the adverse

consequences of misplaced trust.

5.2 Limitations

This study might have several limitations such as:

• The use of the Wizard of Oz technique, while

effective for simulating generative AI functional-

ity, may not fully replicate real-world interactions

with generative AI tools.

• The participants consisted solely of Master’s stu-

dents, potentially limiting the generalizability of

the ﬁndings to other groups with varying educa-

tional backgrounds. Future research could ad-

dress these limitations by including more diverse

participant groups.

• The variation in UI design was limited to the

avatar and text font. Exploring additional design

variables could further enhance understanding of

the relationship between UI design and user trust

in AI systems.

6 CONCLUSION AND FUTURE

PLAN

This study’s ﬁndings highlighted that avatars are the

most inﬂuential UI element affecting user trust in

generative AI. Results showed also that participants

demonstrated sensitivity to text font variations. Inter-

estingly, the results showed that despite participants

interacting with the same source of outputs, variations

in UI led to differing perceptions of trust, emphasiz-

ing the role of UI design in shaping trust in genera-

tive AI responses. These results emphasize the impor-

tance for designers and developers to exercise caution

when designing UI, guiding users to avoid placing

excessive trust in unregulated generative AI systems.

Users should not be misled by UI design choices into

increasing their trust in such systems.

For future research, we plan to conduct a larger ex-

periment with participants from diverse user groups,

varying in educational background and familiarity

with generative AI. We also aim to explore the nu-

anced interactions between UI design and user trust

in generative AI tools by considering additional UI

elements, such as color, which were not a factor stud-

ied in this experiment as the three UI shared the color

blue.

REFERENCES

Alagarsamy, S. and Mehrolia, S. (2023). Exploring chat-

bot trust: Antecedents and behavioural outcomes. He-

liyon, 9(5).

Atillah, I. E. (31 March 2023). Man ends his

life after an AI chatbot ’encouraged’ him

to sacriﬁce himself to stop climate change.

https://www.euronews.com/next/2023/03/31/man-

ends-his-life-after-an-ai-chatbot-encouraged-him-to-

sacriﬁce-himself-to-stop-climate-.

Bach, T. A., Khan, A., Hallock, H., Beltr

ao, G., and Sousa,

S. (2024). A systematic literature review of user

trust in AI-enabled systems: An HCI perspective. In-

ternational Journal of Human–Computer Interaction,

40(5):1251–1266.

Bae, S., Lee, Y. K., and Hahn, S. (2023). Friendly-bot: The

impact of chatbot appearance and relationship style on

user trust. In Proceedings of the Annual Meeting of the

Cognitive Science Society, volume 45.

Baek, T. H. and Kim, M. (2023). Is ChatGPT scary good?

how user motivations affect creepiness and trust in

generative artiﬁcial intelligence. Telematics and In-

formatics, 83:102030.

Feuerriegel, S., Hartmann, J., Janiesch, C., and Zschech,

P. (2024). Generative AI. Business & Information

Systems Engineering, 66(1):111–126.

Fowler, G. A. (10 Aug 2023). AI is acting

‘pro-anorexia’ and tech companies aren’t

Exploring the Inﬂuence of User Interface on User Trust in Generative AI

713

stopping it. [Accessed on April 2024].

www.washingtonpost.com/technology/2023/08/07/ai-

eating-disorders-thinspo-anorexia-bulimia/.

Gupta, A., Basu, D., Ghantasala, R., Qiu, S., and Gadiraju,

U. (2022). To trust or not to trust: How a conver-

sational interface affects trust in a decision support

system. In Proceedings of the ACM Web Conference

2022, pages 3531–3540.

ISO/IEC 25022 (2016). ISO/IEC 25022:2016. Systems and

software engineering – Systems and software qual-

ity requirements and evaluation (SQuaRE) – Measure-

ment of quality in use.

Klar, R. (08 July 2023). AI chatbots pro-

vided harmful eating disorder con-

tent: report. [Accessed on April 2024].

https://thehill.com/policy/technology/4141648-ai-

chatbots-provided-harmful-eating-disorder-content-

report/.

Liu, Y. and Siau, K. L. (2023). Human-ai interaction and

AI avatars. In International Conference on Human-

Computer Interaction, pages 120–130. Springer.

Logan, N. (4 Feb 2024). This article is real

— but AI-generated deepfakes look damn close

and are scamming people. [Accessed on April

2024]. https://www.cbc.ca/news/canada/deepfake-ai-

scam-ads-1.7104225.

McKinsey & Company (25 Aug 2023). What’s

the future of generative AI? an early view

in 15 charts. [Accessed on December

2024]. https://www.mckinsey.com/featured-

insights/mckinsey-explainers/whats-the-future-

of-generative-ai-an-early-view-in-15-charts#/.

Milmo, D. (22 Feb 2024). Google pauses AI-

generated images of people after ethnic-

ity criticism. [Accessed on April 2024].

www.theguardian.com/technology/2024/feb/22/google-

pauses-ai-generated-images-of-people-after-

ethnicity-criticism.

Nowak, K. L. and Rauh, C. (2005). The inﬂuence of

the avatar on online perceptions of anthropomor-

phism, androgyny, credibility, homophily, and attrac-

tion. Journal of Computer-Mediated Communication,

11(1):153–178.

Paul, S. and Rosala, M. (2024). The Wizard

of OZ method in UX. [Accessed on December

2024]. https://www.nngroup.com/articles/wizard-of-

oz/. Nielsen Norman Group.

Sara, R. and Maria, R. (19 April 2024). The Wizard

of Oz Method in UX. [Accessed on April 2024].

https://www.nngroup.com/articles/wizard-of-oz/.

Seitz, L., Bekmeier-Feuerhahn, S., and Gohil, K. (2022).

Can we trust a chatbot like a physician? a qualitative

study on understanding the emergence of trust toward

diagnostic chatbots. International Journal of Human-

Computer Studies, 165:102848.

Sousa, S., Cravino, J., and Martins, P. (2023). Challenges

and trends in user trust discourse in AI popularity.

Multimodal Technologies and Interaction, 7(2):13.

Ueno, T., Sawa, Y., Kim, Y., Urakami, J., Oura, H., and

Seaborn, K. (2022). Trust in human-AI interaction:

Scoping out models, measures, and methods. In CHI

Conference on Human Factors in Computing Systems

Extended Abstracts, pages 1–7.

Walsh, M. (22 April 2024). ChatGPT Statistics — The

Key Facts and Figures. [Accessed on April 2024].

https://www.stylefactoryproductions.com/blog/chatgpt-

statistics.

Yang, R. and Wibowo, S. (2022). User trust in artiﬁcial

intelligence: A comprehensive conceptual framework.

Electronic Markets, 32(4):2053–2077.

Yashmi, N., Momenzadeh, E., Anvari, S. T., Adibzade, P.,

Moosavipoor, M., Sarikhani, M., and Feridouni, K.

(2020). The effect of interface on user trust; user be-

havior in e-commerce products. In Proceedings of

the Design Society: DESIGN Conference, volume 1,

pages 1589–1596. Cambridge University Press.

Yen, C. and Chiang, M.-C. (2021). Trust me, if you

can: a study on the factors that inﬂuence consumers’

purchase intention triggered by chatbots based on

brain image evidence and self-reported assessments.

Behaviour & Information Technology, 40(11):1177–

1194.

Zierau, N., Flock, K., Janson, A., S

ollner, M., and Leimeis-

ter, J. M. (2021). The inﬂuence of AI-based chatbots

and their design on users’ trust and information shar-

ing in online loan applications. In Hawaii Interna-

tional Conference on System Sciences (HICSS).-Koloa

(Hawaii), USA.

ENASE 2025 - 20th International Conference on Evaluation of Novel Approaches to Software Engineering

714