NAME: DEGREE: BA (Hons) Business with Accounting

SUPERVISOR: Kathryn Haynes

TITLE: Exploring the Impact of Artificial Intelligence on the Role of Internal Auditors in
Financial Institutions
DATE: May 2021
CAMPUS: Newcastle STUDENT NO:

Management Enquiry submitted in partial fulfilment of the requirements of the

BA (HONS) Business with Accounting Programme of Northumbria University
 I declare the following:

that the material contained in this Management Enquiry is the end result of my own
work and that due acknowledgement has been given in the references to ALL
sources be they printed, electronic or personal.

1. the Word Count of this Management Enquiry is 7,076

2. that unless this Management Enquiry has been confirmed as confidential, I
agree to an entire electronic copy or sections of the Management Enquiry to
being placed on Blackboard, if deemed appropriate, to allow future students
the opportunity to see examples of past Management Enquiry. I understand
that if displayed on Blackboard it would be made available for no longer than
five years and that students would be able to print off copies or download.
The authorship would remain anonymous.
3. I agree to my Management Enquiry being submitted to a plagiarism detection
service, where it will be stored in a database and compared against work
submitted from this or any other School or from other institutions using the
service. In the event of the service detecting a high degree of similarity
between content within the service this will be reported back to my
supervisor and second marker, who may decide to undertake further
investigation which may ultimately lead to disciplinary actions, should
instances of plagiarism be detected.
4. I have read the University Policy Statement on Ethics in Research and
Consultancy and the Policy for Informed Consent in Research and
Consultancy and I declare that ethical issues have been considered and
taken into account in this research
5. I have read the University Policy Statement on Data Protection in Research
and Consultancy and I declare that the data collected for use in this
Management Enquiry has been properly safeguarded and will be destroyed
once the Management Enquiry or subsequent research activity has been
concluded. I acknowledge that it is my responsibility to destroy the
information with due regard to confidentiality

SIGNED: DATE:18/05/21

1
Student name:
Degree: BA (Hons) Business with Accounting
Supervisor: Kathryn Haynes
Title: Exploring the Impact of Artificial Intelligence on the Role of Internal Auditors in
Financial Institutions

Date: May 2021

Keywords: Internal-Audit, Internal-Auditor, Artificial-Intelligence

Abstract

As global adoption of artificial intelligence is rapidly expanding, internal auditors need

to adapt to ensure they remain relevant. However, previous research has
predominantly focused on the impact of artificial intelligence on external audit and
limited research has been conducted in the context of internal audit. Therefore, this
qualitative study explores the impact of artificial intelligence on the role of internal
auditors in Financial Institutions, to contribute new knowledge by addressing the
current gaps in internal audit literature. The case organisation, Organisation X is a
Financial Institution that offers commercial and retail banking services. Based on
seven semi-structured interviews conducted with internal auditors employed at
Organisation X, perceptions and feelings towards AI were explored. Findings
demonstrated how artificial intelligence has a significant impact on the working
practices of an internal auditor which was also a significant source of anxiety
amongst participants. On this basis, it was concluded internal auditors are not
sufficiently prepared for the implementation of AI. Therefore, practical
recommendations for organisations were created derived from these findings to
facilitate internal auditors’ adaptability to AI.

2
Acknowledgements
I would like to express my deepest appreciation to my supervisor, Professor Kathryn
Haynes for her continuous support, guidance and invaluable advice throughout the
research process. I would also like to thank the Internal Audit Graduate at
Organisation X who facilitated access to the organisation and participants, as without
this Enquiry would not have been possible. My appreciation is also extended to
Organisation X and the seven Internal Auditors who participated in this research for
their time and contributions.

3
Table of contents
Abstract.........................................................................................................................2
Acknowledgments........................................................................................................3

Chapter 1: Engage......................................................................................................6

1.1 Introduction.......................................................................................................................6
1.2 Research Method and Limitation..................................................................................7
1.3 Data Collection.................................................................................................................8
1.4 Sampling and Recruitment.............................................................................................9
1.5 Data Analysis…..............................................................................................................10
1.6 Trustworthiness…..........................................................................................................12
1.7 Ethical Considerations..................................................................................................13

Chapter 2: Engage and Reflect...............................................................................15

2.1 Introduction......................................................................................................................15
2.2 Theme: Impact on Working Practices...................................................................16
2.2.1 Sub-theme: Skills…..................................................................................................17
2.2.2 Sub-theme: Tasks….................................................................................................18
2.2.3 Sub-theme Focus…..................................................................................................18

2.3 Theme: Job Security…..................................................................................................19

2.4 Theme: Emotions and Affect…....................................................................................21

2.5 Theme: Attitudes towards AI.................................................................................22

2.6 Theme: Benefits of AI....................................................................................................24

2.6.1 Efficiency.......................................................................................................................24
2.6.2 Accuracy.......................................................................................................................24
2.7 Discussion.......................................................................................................................25
2.8 Summary…......................................................................................................................26
Chapter 2: Reflect and Connect..............................................................................27

3.1 Drawing Conclusions….................................................................................................27

3.2 Conclusions and Limitations…...................................................................................29

4
3.3 Recommendations........................................................................................................29
3.4 Action Plan.....................................................................................................................32
3.5 Future Research….......................................................................................................35
References…...................................................................................................................36-40
Appendices......................................................................................................................41-48

Glossary:
IAF- Internal Audit Function
IA- Internal Audit
AI- Artificial Intelligence

5
Introduction

This Enquiry is aiming to explore the impact of AI on the role of internal auditors. The
literature review in Report 1 identified impacts of AI, predominantly on external audit
due to limited research in the context of internal audit (IA) and highlighted several
gaps in literature. In this report, the qualitative methods used in this Enquiry are
introduced, in addition to the method of data collection, sampling, and analysis
process. The next chapter presents the key findings, that are analysed and
synthesised with the literature reviewed in Report 1 to establish the relations
between the findings and existing knowledge. Finally, conclusions are drawn from
the key findings and practical recommendations are presented, alongside an action
plan for Organisation X to facilitate implementation.

Chapter 1: Engage (Methods)

1.1 Introduction

The purpose of this chapter is to introduce the research methods used for this
Enquiry and evidence the suitability of adopting qualitative methods to explore the
impact of AI on the role of internal auditors to answer the research questions. Firstly,
the chapter discusses qualitative methods, including limitations followed by a critical
justification for selecting semi-structured interviews for data collection. Thereafter,
details of the design and practical implementation are provided, including the
sampling process and data analysis. Finally, ethical issues are considered per
University Guidelines.

Research Questions:

RQ1- How do internal auditors perceive the use of AI in internal auditing?

RQ2- What are internal auditors’ perceptions and feelings towards how AI will impact
their role?

6
1.2 Research Method and Limitations

Previous studies relating to the topic primarily use quantitative methods (e.g., Afroze
& Aulad, 2020; Whitman & Sobczack, 2018), enabling the researchers to establish
cause-effect relationships between variables and test hypotheses (Ahmad, Wasim,
Gogoi, & Ifran, 2019). However, this study adopted qualitative methods which allow
for an understanding of behaviours, beliefs and values (Choy, 2014) to gain an in-
depth insight into internal auditor’s perceptions and feelings towards AI, which are
strengths of this method (Mohajan, 2018). Qualitative methods aim to address the
‘what’ and ‘how’ of a phenomenon (McCusker & Gunaydin, 2014) and are best
employed to answer questions about “experience, meaning and perspective” from
the participants’ viewpoint (Hammarberg, Kirkman and Lacey,2016, p. 499). This
evidences the appropriateness of using qualitative methods to explore how internal
auditors perceive the phenomenon of AI, therefore providing rationale for this choice
due to suitability in addressing the research questions. Qualitative research also
produces detailed information which as a result can facilitate a deeper understanding
of the research phenomenon (Querios, Faria, & Almeida, 2017; Cresswell, 2014),
highlighting a strength in comparison to quantitative. Despite their strengths,
qualitative methods are criticised by scholars (e.g., Mohajan, 2018; Alshenqeeti,
2014) as they are not statistically representative due to requiring smaller sample
sizes and therefore results are limited in terms of generalisability. However, as
qualitative research aims to explore individual experience and achieve depth of
understanding rather than breadth (Patton, 2002), generalisability is not intended.
Therefore, qualitative methods are more appropriate for the purpose of this study.

1.3 Data Collection

To obtain perceptions and feelings towards AI, interviews were used for data
collection as the method aims to elicit the views of participants and access their
feelings, according to Bloomberg and Volpe (2019). Interviews are considered
fundamental in qualitative research (Knox & Burkard, 2008), being the most common
method of qualitative data collection (Jamshed, 2014; Cassell, 2005). However,
interviews are rarely utilised for data collection in IA research, with surveys and other
qualitative methods dominating the current literature (Kotb, Elbardan, & Halabi,

7
2020). Therefore, by selecting interviews this study aims to address this gap.
Individual interviews were preferred over other qualitative methods such as focus
groups as they produce more detail and enable a deeper insight, due to their
personalised approach (Abhabi & Anozie, 2017). The interviews were semi-
structured, meaning predetermined, open-ended questions were asked in addition to
further probing questions that arose from interviewee responses (Roulston & Choi,
2017). This produced rich data enabling greater depth into perceptions (McIntosh &
Morse, 2015; Peters & Halcomb, 2015). This degree of flexibility also provided
participants with the opportunity to “offer new meanings” to the Enquiry focus
(Galletta, 2013, p. 24), while ensuring the necessary aspects of the research
questions were still addressed, which may not have occurred using an unstructured
approach. As a result, new concepts emerged which Mohajan (2018) notes as a
strength of the method. Nevertheless, the extent to which flexibility is an advantage
is disputed by Fox (2009), who highlights difficulties regarding inconsistency in
responses due to the lack of standardisation. This is substantiated by Hofisi, Hofisi
and Mago (2014) who explain that as a result, concerns over dependability may
occur. Conflicting this, Saunders, Lewis and Thornhill (2019) note how semi-
structured interview findings are not intended to be repeatable and adopting a
structured approach would have restricted the scope of the findings in this study.

The interview schedule (appendix B) consisted of 10 questions, grouped into themes

that were developed using the research questions as a framework and informed by
the conceptual framework developed from the literature review. Preceding this,
participants were provided with an overview of the research and asked questions
relating to their experience and department to obtain demographic information and
build rapport. Rapport is essential for effective communication can improve the
quality of data obtained in qualitative research (Zakaria & Mustamal, 2014; DiCicco-
Bloom & Crabtree, 2006). The interviews were synchronous electronic interviews,
arranged via email and computer-mediated using Zoom due to government
restrictions. Unlike telephone or asynchronous interactions, this format replicates
elements of face-to-face interviewing and therefore maintains similar advantages,
such as the ability to recognise and respond to cues (Lolacono, Symonds, & Brown,
2016). Moreover, further advantages were apparent such as enhanced accessibility
to the sample through removing geographical constraints as Organisation X is

8
situated in a different country, achieving time and cost-effectiveness also
(Janghorban, Roudsari, & Taghipour, 2014). Zoom is considered a viable tool for
qualitative data collection (Archibald, Ambagtsheer, Casey, & Lawless, 2019) and
was selected due to participants’ preference, which circumvented weaknesses
related to participants’ technical abilities (Gray, Wong-Wylie, Rempel, & Cook, 2020).
However, on occasion, minor technical difficulties such as connectivity issues were
experienced, which is a considerable weakness of the method (Weller, 2015).

Between March 9th-16th 2021, seven interviews were conducted lasting a duration of
14 to 26 minutes each. With prior consent, the interviews were audio-recorded using
a laptop, allowing valuable details to be captured, and then transcribed verbatim
following the interview to maximise data accuracy, despite this being time-consuming
(Tessier, 2012). Data was also recorded using notes that were summarised into key
points and then discussed with participants. This clarified any misperceptions
regarding the questions asked or responses given, which may occur during
interviews due to ambiguities in language (Alshenqeeti, 2014; Atieno, 2009),
therefore increasing credibility.

1.4 Sampling and Recruitment

To enable an understanding of how Internal Auditors at Organisation X perceive AI

and achieve objective 2, a purposive sampling strategy was employed to select
information rich individuals knowledgeable about the topic (Patton, 2002). A non-
probability sampling approach is often criticised due to the size and lack of
representativeness of the sample (Albert, 2010). However, inclusion criteria required
participants to be an Internal Auditor employed at Organisation X to ensure
relevancy to the case organisation. Moreover, as guidelines for non-probability
sample sizes are “virtually non-existent” (Guest, Bunce, & Johnson, 2006, p. 59), the
sample size was determined by University guidance and enabled sufficient data to
be obtained. Access was obtained through a gatekeeper, who facilitated recruitment
by emailing prospective participants, which was beneficial due to limited access to
the research site (McFayden & Rankin, 2016). The sample consisted of seven
Internal Auditors with around four years’ experience working across five different
departments (table1).

9
Table 1 Sample Characteristics

Participant Gender Experience (Years) Department

1 Female 0-1 Infrastructure and Cyber Security
2 Female 2-3 Retail Banking
3 Male 1-2 Markets
4 Female 0-1 Finance and Treasury
5 Male 1-2 Financial Crime
6 Male 3-4 Retail Banking
7 Male 4+ Retail Banking

1.5 Data Analysis

To identify prominent, recurring themes in the data to achieve objective 3, thematic

analysis was used in this study. Thematic analysis enables flexibility (Javadi &
Zarea, 2016), generating rich, unanticipated insights as a result (Nowell, Norris,
White & Moules, 2017; Staller, 2015). According to Braun and Clarke (2006),
thematic analysis is particularly useful for examining and comparing the perspectives
of participants, providing further justification for thematically analysing the data due
to relevance to the research questions. An inductive approach was adopted,
meaning themes were derived from the data and analysis was conducted according
to the six phases of thematic analysis developed by Clarke and Braun (2006).

10
Table 2 Phases of Thematic Analysis Followed

Phase Description of process

1. Familiarise yourself • Transcribe and re-read the data
with data • Note down any initial ideas
2. Generate initial • Code interesting features in a systematic
codes fashion across entire data set
• Collate data relevant to each code
3. Search for themes • Collate codes into potential themes
• Gather relevant data for each theme
4. Review themes • Check if the themes work in relation to coded
extracts and the entire data set
• Generate a thematic map of analysis
5. Define and name • Continual analysis refining the specifics of
themes each theme and overall story analysis tells
• Generate clear theme definitions and names
6. Produce report • Select vivid and compelling extract examples
• Analyse selected extracts in relation to
research question and literature
• Produce a report of the completed analysis

Following phase 1, transcripts were highlighted by hand and codes were generated,
capturing interesting features of the data relevant to the research questions (Maguire
& Delahunt, 2017). This accentuated recurrent topics and meanings in the data
(Cope, 2010), relevant to objective 3. During phase 3, codes that fit together were
combined into potential themes (appendix C), which were then refined to ensure that
themes accurately represented and were supported by sufficient data (Braun &
Clarke, 2006). During refinement, it was decided that codes reflected two main
Benefits of AI, therefore this theme was separated into two subthemes, Efficiency
and Accuracy. Moreover, the theme of Challenges was discarded due the challenge
identified fitting better within Skills. Once themes were renamed, a thematic map was
created (figure1) and themes were defined ready for written analysis.

11
Figure 1 Thematic Map

1.6 Trustworthiness

Scholars believe evaluating the trustworthiness of qualitative research requires

different criteria than quantitative, and involves establishing credibility, transferability,
dependability and confirmability according to Lincoln and Guba (1985).

Credibility defined by Bloomberg and Volpe (2019) concerns the extent participants’
views correspond with their portrayal by the researcher to demonstrate truth in
research findings. Issues relating to credibility were addressed throughout the study,
through the clarification of meanings and strategies were employed to bolster
credibility. As the researcher is the main instrument in qualitative research, Wa-
Mbaleka (2019) argues it is vital their position is considered throughout. Reflexivity,
which refers to awareness of the researcher’s role and how this impacts the research
process (Haynes, 2012) was essential to achieve this. In this study, the researcher’s
position as a woman of similar age and level meant rapport was easier to build and
participants were more open, evidenced by one who disclosed that they would not
say the same to management. This is because commonalities in facets such as

12
gender and age between a researcher and participant make both parties more
comfortable, resulting in a more honest discussion (Manderson, Bennett, & Andajani-
Sutjahjo, 2006). Therefore, overlap in facets can be beneficial during research
interactions (Jacobson & Mustafa, 2019). Dependability is the extent research
processes are documented to enable consistency and stability of findings (Moon,
Brewer, Hartley, Adams, & Blackman, 2016). Concerning this study, dependability is
demonstrated through the audit trail, detailing data collection and analysis, in
addition to examples of transcriptions and coding. Confirmability relates to findings
and interpretations being derived from data rather than researcher bias (Connelly,
2016). This is evidenced through justifications of the data collection and analytical
methods presented in this chapter, as well as the reflexivity process. Transferability,
referring to the extent findings apply to other contexts was improved by adopting a
purposeful sampling strategy (Bloomberg & Volpe, 2019).

1.7 Ethical Considerations

When conducting social science research, it is vital ethical issues relating to

informed consent, harm, invasion of privacy and deception are considered (Bryman
& Bell, 2015). Obtaining informed consent involves providing sufficient information
about aspects of the study so participants can make an informed decision to consent
before participating (Nijhawan, et al., 2013). In addition to organisational consent
(appendix D), informed consent was gained prior to the study (appendix E) as
participants were informed of the research purpose to avoid deception (Tai, 2012),
alongside how data would be stored and could be withdrawn (appendix F).
Additionally, it was explained how anonymity would be maintained, which was done
through referring to participants by their unidentifiable code throughout reporting,
ensuring confidentiality (Kaiser, 2009). Furthermore, participants were not subjected
to sensitive questioning as questions remained relevant to the research questions.
Participants were able to refrain from answering if they wished, to protect privacy

13
and reduce the risk of harm (Bryman & Bell, 2015). See table 3 for detail on how
risks were mitigated throughout the process.

Table 3 Ethical Risks and Mitigations

14
Chapter 2: Engage and Reflect

2.1 Introduction

This section comprises of both the Engage and Reflect Chapters of the Management
Enquiry, appropriate due to the qualitative approach adopted in this Enquiry to
explore the impact of AI on internal auditors. Firstly, the chapter provides a
description of the thematic analysis method employed, including the themes and
sub-themes derived from the data (table4). Secondly, findings are presented in order
of significance, then analysed and interpretated using the concepts identified in
Report 1. This is followed by an overarching discussion summarising the findings in
relation to the research questions and existing knowledge.

Clarke and Braun (2017) describe thematic analysis as a method to identify,

analyse and interpret patterns of meaning or themes from data that is used
extensively in qualitative research (Castleberry & Nolen, 2018). Data was analysed
thematically following the phases outlined in Chapter 1 and five key themes in
addition to five sub-themes were derived from the analysis.

15
 Table 4 Themes and Sub-themes

2.2 Theme: Impact on Working Practices

This theme explores internal auditors’ perceptions regarding how AI will impact
working practices, fundamental in answering RQ2. Overall, participants
demonstrated consensus in the view that AI would not alter their role, rather how the
role is carried out as one interviewee emphasised:

“Mainly our role is to provide independent, objective assurance to the bank’s control
environment and with AI, I don’t think our role will change, but I think how we go
about executing that role definitely will.” (P7)

Therefore, the following sub-themes explore perceptions of the three main impacts in
more detail.

16
2.2.1 Sub-theme: Skills

One main topic discussed in the study related to skills, participants were unanimous
in the view that technical skills will be essential following the implementation of AI as
P6 explained:

“You’ll need the specifics, so knowing how to code, use data, python and
whatever so my skill set definitely needs to change so I can use that stuff”

Similarly, P7 agreed:

“There will be a need to understand the technical element of AI, and if we’re using AI
to perform audit work clearly there will need to be an evolution in skillset to be able
to do that”

The reoccurring focus on skills by participants demonstrates a significant impact of

AI on internal auditors to be a change in skill demand, evidencing the necessity for
organisations to upskill internal auditors in AI. While IA research has previously
failed to explore this, these findings are consistent with external audit literature
(Afroze & Aulad, 2020; Issa, Sun & Vasarhelyi). Moreover, these findings mirror the
impact of data analytics in IA (Betti & Sarens, 2020; Tang, Norman, & Vendryk,
2017).
Therefore, this finding addresses the gap in the literature by contributing an
understanding of how AI impacts the skills of an internal auditor. Nevertheless,
concerning existing practice, participants revealed they had received minimal
training to develop AI skills, despite recognising the requirement to. Therefore, this is
worth significant consideration as organisations should seek to evolve skills, as
failure to do so will decrease AIs perceived ease of use, subsequently increasing
anxieties and reducing acceptance also (Davis, 1989). Consequently, 4/7
participants anticipated their role will merge with a data analyst as P6 emphasised:

“The direction of travel is definitely towards a hybrid of a data analyst and internal
auditor.”

Similarly, P5 stated:

“I think you’ll basically need to be a data analyst to understand how it works”

17
This demonstrates that following the implementation of AI, the role of an internal
auditor may evolve towards that of a data analyst which is consistent with the view of
the Institute of Internal Auditors (2020). This is significant to Organisation X, as
increasing focus on technical skills will ensure internal auditors are adequately
prepared for this.

2.2.2 Sub-theme: Tasks

A recurrent topic discussed by participants related to the impact of AI on tasks. In

line with prior research, all participants anticipated that AI would replace their tasks
through automation, as one interviewee referring to mortgage price changes
commented:

“It’s going into a direction where eventually AI will be able to key in and make those
changes without any manual interventions.” (P2)

Comparably, another stated:

“It can easily automatically extract data which will reduce the need for manual
extraction and take a way a lot of the manual checking and testing that we do so our
role will definitely change in terms of this.” (P1)

This finding indicates that all participants agree through the automation of tasks, AI
will replace aspects of their role, which supports the stance of scholars who argue
this a main impact of AI in auditing (Kokina & Davenport, 2017; Issa, Sun &
Vasarhelyi, 2017). Unexpectedly, this may also increase internal auditors’ motivation,
as participants explained how completing manual tasks such as ticking is often
demotivating. Therefore, these findings extend existing knowledge by explaining a
benefit of this impact.

2.2.3 Sub-theme: Focus

Although all participants agreed AI would not alter the role or objectives of the IAF,
some participants (2/7) mentioned that AI may impact the focus as one explained:

“I think it will let us focus more on what we already do like providing assurance” (P5)

Likewise, another commented:

18
“Our objectives as function won’t change but I think using AI will increase our focus
on specific areas” (P3)

This suggests while participants do not believe the role of the IAF will change, they
anticipate an increase in focus on assurance activities. Interestingly, the finding
presented directly conflicts Betti and Sarens (2020) who postulate that the IAF will
adopt the role of a partner rather than a controller, influenced by the increasing
demand for consulting activities due to technology. The dissimilarity in the results
may have occurred due to organisational context, as the level of involvement in
assurance vs consulting activities differs between firms (Soh & Martinov-Bennie,
2011; Selim, Woodward & Allegrini, 2009). Additionally, the polarity could be a result
of participant experience, as Betti and Sarens (2020) interviewed chief audit
executives and senior management, whereas this study consisted of graduates,
therefore perspectives may have been impacted by experience. Alternatively, as the
literature lacked specificity in terms of which technologies, it could be argued the
impact was due to other technologies rather than AI, thus providing a further
plausible interpretation. This finding contributes new insights to literature regarding
how AI may impact the role and activities of the IAF in Financial Institutions.
However, the applicability of the findings may be limited by organisational context
due to differing levels of involvement in activities and Organisation X being a
Financial Institution. Therefore, although beyond the scope of this study due to
relevance to the case organisation and research objectives, future research could
explore the impact of AI on the role of the IAF in other organisations, in addition to
Financial Institutions.

2.3 Theme: Job Security

This theme explores participants views on job security, as following discussion

concerning the automation of tasks, participants (3/7) communicated concerns over
being replaced, as a result as P4 speculated:

“It will be able to do it for me which will make us less in demand and less needed in
our jobs” later adding “what will be the need for us to actually even do our work”.

Another mentioned:

19
“It will mean a human being won’t be needed to do those things. But personally, I
think that’s quite a while off.” (P2)

This demonstrates that some internal auditors believe the implementation of AI may
lead to their role being replaced, which aligns with Caratas, Spatariu and Gheorghiu
(2018) who postulate that AI will eventually replace internal auditors. Additionally,
this is substantiated by the findings of Frey and Osborne (2013), and Bakhashi, Frey
and Osborne (2015), relating to the probability of computerisation for the auditing
profession. This finding is significant as uncertainty over job security caused
anxieties and dislike towards AI amongst participants. In contrast, however, this
notion was rejected by some participants who argued against AI’s ability to fully
replicate human judgement and subjectivity as reason:

“My role in financial crime requires a lot of judgement and there’s nothing really at
this moment in time that suggests AI has the capability to do that” (P5)

The lack of consensus demonstrated by participants mirrors the findings of Lois,

Drogalas, Karagiorgos and Tsikalakis (2020) and current disagreement in IA
literature regarding the extent of impact AI will have and whether the requirement for
internal auditors will be replaced. A possible cause for the disagreement in findings
and current literature may relate to inconsistency in the time frame that participants
are referring to. This is exemplified by one interviewee who believed the impact will
be limited in the shorter term but expressed doubts relating to the long term:

“In the short to medium term there is always going to be the need to do that work,
but I don’t know about how it will be in 20 plus years’ time because that’s any bodies
guess.” (P7)

It can therefore be argued that although AI isn’t expected to replace auditors in the
short-term, it is unknown whether this will occur long-term. While this finding was
unable to confirm whether AI will replace internal auditors, it adds weight to
existing arguments and provides a clearer understanding of why perspectives on
this topic may vary. Future research could consider specific time frames to achieve
clarification.

20
2.4 Theme: Emotions and Affect

This theme explores internal auditors’ emotions and feelings towards AI, relevant
to RQ2. Researchers found that the implementation of AI resulted in negative
feelings among external auditors. In this study, a significant finding was that 4/7
participants reported negative feelings towards AI as one interviewee explained:

“I think it’s daunting for me because I’m not tech savvy at all.” (P2)

Similarly, another explained:

“I don’t come from a techy background, so I’m just a bit lost with it” (P6)

This clearly illustrates that some internal auditors feel apprehensive towards the use
of AI, which is significant as previous literature has failed to explore this. Therefore,
this finding allows a new understanding of internal auditors’ feelings towards AI and
contributes to answering RQ2. Although there is insufficient research for comparison
in the context of IA, this is consistent with external audit literature in Report 1.
Specifically, this finding resonates with Whitman and Sobczack (2020) who
reported anxieties in external auditors, primarily caused by the increased risk
associated with AI. The finding in this study however could be attributed to
participants experience with AI, as all 3 participants indicated that this was limited
due to their educational background and work experience as P2 explained:

“I’ve never really had exposure to it so maybe people who did a technology focused
degree wouldn’t struggle as much as me”

It can therefore be suggested that internal auditors with less AI experience are more
likely to perceive it as difficult to use and experience negative feelings towards it.
This builds on the research of Whitman and Sobczack (2020) by providing a
possible alternative explanation for their results. This notion is substantiated by
existing theory relating to technology acceptance (1989), and information technology
acceptance in internal auditors (2009), as less relevant experience, decreases a
technology’s perceived ease of use, thus increasing anxiety. Consequently, a way
to improve feelings towards AI at Organisation X would be to increasing involvement
in AI. P5 also illustrated the effectiveness of this:

21
“I’m becoming more involved in that type of work so now I don’t feel too bad, but
when I started, I was very very apprehensive and not too keen on it at all.”

This echoes the findings of Lai, Dai, Gershberg, & Vasarhelyi (2018), who
established that as internal auditors gained knowledge and experience using audit
analytics, their feelings improved simultaneously. This is worth significant
consideration for Organisation X and in practice, as it explains why internal auditors
may feel apprehensive towards AI and proposes a possible solution to address this.

2.5 Theme: Attitudes towards AI

This theme discusses internal auditors’ attitudes towards AI, significant in answering
RQ1. Overall, the majority of participants (5/7) responded positively as one
interviewee mentioned:

“It’ll definitely be a good thing for us.” (P1)

Another emphasised,

“I think internal audit really needs to embrace it.” (P7)

However, a few participants (2/7) exhibited negative attitudes as the comment below
firmly illustrates:

“Generally, I think I’m quite anti-AI and personally I don’t actually like it at all.” (P4)

The dissimilitude in responses demonstrates that although attitudes were

predominantly positive, this was not accurate for all and therefore this view cannot
be generalised. This provides an original insight into internal auditors’ attitudes
towards AI, as previous research has failed to explore this. One possible reason for
the dissimilarity may relate to the extent participants perceive AI to be relevant to
their role as P4 commented:

22
“I don’t know how” later adding “I think it’s getting to the point where it’s actually a bit
unnecessary.”

Whereas in contrast, P7 who responded positively and believed AI is “necessary”

stated:

“AI is definitely relevant and is only going to actually become more relevant”

This illustrates that attitudes towards AI may have been influenced by an

understanding of its relevance to their role. This corroborates the ideas of Kim,
Mannino and Nieschwietz (2009), who posit job relevance as a significant factor
influencing internal auditors’ perceptions of information technology. Seminal work
(Davis, 1989; Venkatesh and Davis, 2002) substantiates this argument, by
explaining how the extent technology is considered relevant impacts its perceived
usefulness and ultimately its acceptance. Hence, it can be inferred that internal
auditors who perceive AI as relevant are more likely to adopt a positive attitude
towards it. Thus, to improve attitudes towards AI, organisations should seek to
increase employees understanding of its relevance. This finding is significant as it
provides an understanding as to why internal auditors may express dislike towards
AI. Additionally, dislike may have also been related to participants concerns over job
security as when referring to this P4 confessed:

“I guess that’s why I’m a bit resistant towards it”

Therefore, this provides an alternative explanation for the finding in this study that is
also plausible. This is relevant as it provides a deeper insight into factors influencing
internal auditors’ attitudes towards AI, which contributes to research objective 4,
relating to assessing variables influencing perceptions of AI.

23
2.6 Theme: Benefits of AI

This theme explores Efficiency and Accuracy, the two main benefits of AI identified
by participants and is relevant in answering RQ1.

2.6.1 Sub-theme: Efficiency

A prominent view related to efficiency, as participants consistently emphasised AI’s

ability to save time and achieve greater coverage:

“It’ll speed up testing, meaning we’re more streamlined” (P5)

And:

“AI will enable us to access and test entire populations rather than samples” (P1)

This finding demonstrates that AI is perceived to be beneficial in IA through

increasing efficiency. While there is limited research for comparison in the context of
IA, due to the gap in the literature identified in Report 1, this finding is consistent with
scholars who propose this as a significant benefit in external auditing (Xing, Lijun, &
Zhu, 2020; Raschke, Saiewitz, Kachroo, & Lennard, 2018). Therefore, through
addressing this gap, this finding contributes new knowledge to IA literature regarding
the benefit of AI.

2.6.2 Sub-theme: Accuracy

In all cases, participants expressed the notion that AI would increase audit accuracy
in either one of two ways, through reducing human error or identifying areas of risk
that would have been missed manually:

“There were probably human errors as I probably made mistakes so I think definitely
AI could have just picked up the problems for me.” (P4)

And,

“AI will make us focus on specific areas as I think it might identify risks that we
wouldn’t be able to find ourselves.” (P6)

These comments regarding AI’s ability to decrease error and find additional risks
evidence perceptions of output quality, an extended factor of TAM that refers to
beliefs that a system would be able to perform tasks well (Venkatesh & Davis, 2000).

24
Previous research (Kim, Mannino and Nieschwietz, 2009; Davis, 1989)
demonstrates that perceptions of high output quality increase a technology’s
perceived usefulness. Therefore, this finding is significant as it indicates that internal
auditors perceive AI as useful due to its output quality. Additionally, it can be
discerned from this finding that participants perceive AI to be beneficial in terms of
enhancing accuracy. In Report 1, it became apparent that limited research has
previously considered this area. However, the level of consensus demonstrated by
participants strengthens the proposition that AI is perceived to increase accuracy in
IA, which reflects the findings of Ding, Lev, Peng, Sun, and Vasarhelyi (2020) and
Hooda, Bawa, & Rana (2018) in the context of external audit.

2.7 Discussion

The overarching aim of this Enquiry was to explore the impact of AI on the role of
internal auditors to answer the following questions:

RQ1: How do internal auditors perceive the use of AI in internal auditing?

RQ2: What are internal auditors’ perceptions and feelings towards how AI will impact
their role?

RQ1

Concerning how AI is perceived in IA, overall, this was predominantly positive as

participants emphasised the relevance of AI, in addition to citing benefits such as
increased accuracy and efficiency due to automation. Although previously
unexplored in relation to IA, these findings are consistent with external audit
literature, therefore this Enquiry extends existing knowledge to the context of IA.
Contrasting Report 1 however, challenges of AI were rarely mentioned by
participants, whereas these were prevalent in the literature reviewed. Nevertheless,
this can be attributed to the fact AI has not been implemented into Organisation X
yet, therefore more challenges may be encountered once this occurs.

25
RQ2

With reference to the conceptual framework developed in Report 1, existing

literature identified three primary impacts of AI on the role of internal auditors, in
terms of their tasks, skills and the function. Perceptions in this study support existing
knowledge regarding how AI impacts tasks through automation and the lack of
consensus whether this will eventually replace the role of an internal auditor.
However, this Enquiry extends existing knowledge by explaining reason for the
conflict in findings. It was also evident in this study that AI increases the demand for
technical skills with is consistent with existing knowledge, relating to external audit.
Therefore, this Enquiry adds to existing theory by evidencing this as an impact in IA
also. Unlike previous research, the qualitative nature of this Enquiry enabled insight
into internal auditors’ feelings towards AI, which were found to be mainly
apprehensive.
Significantly, it was found that auditors’ feelings were primarily influenced by
perceptions of its impact on working practices. This is exemplified by those who were
apprehensive also expressing uncertainties over job security due to the threat of
automation and concerns about not being able to use AI due to the skills required.
Therefore, this Enquiry provides a novel insight into internal auditors’ feelings
towards AI. Through addressing the gaps in the literature to answer the research
question, this Enquiry contributes an original interpretation of internal auditors’
perceptions and feelings towards AI.

2.8 Summary

Through analysing and interpretating the key findings in this Enquiry, several impacts
of AI on the role of internal auditors have been explored in this chapter. One of the
most significant findings to emerge is the impact of AI on working practices, as
through analysing the interrelationships between themes, it was discovered this had
subsequent negative impacts on perceptions and feelings towards AI. Moreover, the
analysis in this chapter enabled a deeper insight into internal auditors’ perceptions of
AI and contributed a new understanding of feelings towards AI through answering
the research questions, thus addressing this gap. This discussion supports the
following chapter to provide practical recommendations derived from these findings
to facilitate internal auditors’ adaptability to AI.
26
Chapter 3: Reflect and Connect

3.1 Drawing conclusions

The aim of this Enquiry was to explore the impact of AI on the role of internal
auditors. Firstly, this chapter will demonstrate the Enquiry’s success in achieving this
aim through fulfilling the research objectives, followed by the key conclusions derived
from the findings. Limitations will be discussed, then recommendations to
Organisation X will be presented, including an action plan. Finally, implications and
areas of future research will be considered.

Objective 1. Examine current knowledge on topic through critically reviewing

literature

This was fulfilled in Report 1 following an exploration of contemporary literature

relating to AI and IA. This provided an understanding of current knowledge on the
topic, accentuating areas of significance that were summarised into a conceptual
framework. This framework was used to develop interview questions and in analysis,
enabling a greater level of synthesis between this study’s findings and existing
knowledge. The critical nature of the review also identified several gaps in literature,
from which the research questions were discerned with the aim of addressing these
to contribute new knowledge to IA literature.

Objective 2. Conduct qualitative interviews to identify the current role of

internal auditors and gain an in depth understanding of perceptions of AI and
its impact on their role

After employing purposive sampling to select individuals knowledgeable about the

topic, seven semi-structured interviews were conducted with Internal Auditors
employed at Organisation X, as discussed in Chapter 1. The interviews explored
participants current role, in terms of their working practices and role as a function,
alongside how they anticipate AI will impact this. The qualitative nature of the
interviews enabled an in-depth understanding of perceptions and feelings towards
AI, in addition to its impact on their role.

Objective 3. Analyse data thematically to identify prominent themes that

reoccur

27
The data was analysed thematically following the phases of thematic analysis
outlined in Chapter 1. Through this process, five prominent, recurring themes and
five sub-themes that captured interesting aspects of the data relevant to the
research questions were identified and presented in Chapter 2. These themes
illustrate main impacts of AI, such as on working practices and the IAF, as perceived
by participants.

Objective 4. Assess external variables influencing perceptions of AI

Through exploring the interrelationships between themes and participant responses,

in addition to drawing on the conceptual framework developed in Report 1, external
variables influencing internal auditors’ perceptions of AI were identified and
discussed in Chapter 2.

Objective 5. Formulate recommendations to management to facilitate internal

auditors’ adaptability to AI

Through analysing the key findings and variables influencing perceptions of AI, three
recommendations have been created to facilitate internal auditors’ adaptability
through developing skills and improving attitudes towards AI. These are presented in
Chapter 3.

28
3.2 Conclusion and Limitations

Given the impact of AI on external audit and limited research in the context of IA
established in Report 1, the overarching aim of this Enquiry was to explore the
impact of AI on the role of internal auditors. Through employing qualitative methods,
several impacts of AI were explored in Chapter 2 and from the key findings
presented, it can be concluded that at present, internal auditors are not adequately
prepared for the implementation of AI or its impact on their role. Therefore,
recommendations have been derived from these findings to facilitate adaptability
and enable internal auditors to prepare for AI.

A limitation of this study relates to demographics of the sample, which consisted of

Internal Auditors with around four years’ experience. This was due to time
constraints and limited availability of more experienced auditors in senior positions.
As discussed in Chapter 2, experience may influence perceptions of AI, therefore
this may limit the extent perceptions reflect those of Internal Auditors across
Organisation X. An extension of this study could involve senior staff to gain
additional perspectives and explore the extent experience influences perceptions of
AI.

3.3 Recommendations

In this section, three practical recommendations derived from the findings are
presented. Benefits are explained, and cost or other resource requirements
considered. Potential implementation challenges are also discussed, alongside how
these may be overcome.

High Priority Recommendation 1: Training

It is recommended that Organisation X provide AI training for Internal Auditors. This

is derived the key finding that majority of participants do not have sufficient skills or
knowledge to use AI and have received minimal training. This presents an issue as
analysis in Chapter 2 established a significant relationship between a lack of AI
knowledge and experiencing negative perceptions or feelings towards AI. Therefore,
this recommendation is a key priority that will be of substantial benefit as training will
increase Internal Auditors’ knowledge of AI, subsequently improving their

29
perceptions and acceptance also, which is supported by prior research (Kim,
Mannino, & Nieschwietz, 2009). Training is also beneficial as participants explained
an AI angle to an audit may be missed due to lack of knowledge. Therefore, using AI
could identify additional risks and save time, thus improving accuracy and efficiency.
This will also enhance the level of protection provided to the bank’s controls and
safety measures, demonstrating significant organisational benefits. Additionally, the
impact of AI training has proven “considerable” in organisations such as Deloitte
(2019, p.7), who run internal AI programmes for employees. Although internal
training is the cheaper option, this may not be practical due to the level of AI
expertise within the organisation and degree of training required by employees.
Therefore, once this is evaluated, it may be decided using an external training
provider is necessary instead. Although more costly, ranging from £950 per
employee for a beginner session to £2,000 for a specialist (JBI Training, 2021),
sessions can be customised depending on learning requirements, which is useful as
skills differed amongst participants.

Objections may occur due to the cost of using an external training provider.
However, as being able to use AI would be beneficial in terms of increasing
efficiency and productivity, this can reduce operating costs, resulting in cost savings
(McKendrick, 2017). Additionally, external training can be more effective than
internal, due to the professional trainers teaching ability and specialist knowledge
(HR News, 2018). Nevertheless, external training may not be necessary depending
on employee requirements.

High Priority Recommendation 2: Meeting

As discussed in Chapter 2, uncertainties surrounding AI and extent of impact it will

have was a significant cause of anxiety and negative perceptions amongst
participants. Consequently, these perceptions and feelings are also likely to reduce
acceptance of AI. Moreover, anxiety due to uncertainty and the threat of automation
can demotivate employees and decrease productivity in the workplace (Darino,
Hancock, & Lazaroff-Puck, 2019), presenting an issue for organisations. One way AI
related anxieties were eased at PwC was through a discussion designed to provide
reassurance and recognise employee concerns (Fisher, 2019). It is therefore
recommended that management adopt this approach and hold a meeting with

30
employees. The purpose is to share relevant information whilst providing employees
the opportunity to voice their concerns and ask questions. This will be beneficial in
terms of reducing uncertainty, which will subsequently decrease anxieties and
improve perceptions towards AI, thus increasing acceptance also.

As there is no consensus regarding the definite impact AI will have, this

recommendation may be objected due to the perceived inability of answering
questions or solving concerns. However, Cote (2019) explains as this may not be
possible, recognising the unknown is necessary. Therefore, management should
instead aim to understand concerns and respond to the best of their ability. As
availability of staff may also be an issue, it is advised the session is hosted via Zoom
to enhance accessibility and enable recording. Key points can then be transcribed,
and the document distributed by email so those unable to attend can access when
convenient.

Recommendation 3: Buddy System

It is recommended that non-technical internal auditors are allocated a technical

internal auditor “buddy” who they can contact if they require help. This relates to the
finding that some Internal Auditors at Organisation X are more knowledge of AI
depending on their department. This recommendation is derived from current
practice at PwC, whereby AI proficient employees informally mentor their peers to
share knowledge and provide help when necessary (Forbes, 2019). This
recommendation is beneficial as organisational factors such as support have
previously been proven to increase internal auditor acceptance of various
technologies (Lai, Dai, Gershberg, & Vasarhelyi, 2018) Additionally, using one-to-
one interactions for support, such as buddying is recommended in practice as a form
of learning to upskill employees in AI (Deloitte, 2020). Not only is this
recommendation beneficial in terms of facilitating knowledge and skill development,
informal learning can also improve employee engagement and motivation (Gardener,
2020), demonstrating additional benefits.

No financial resources are required; however, difficulties may occur due to time and
availability issues as staff may sometimes be too busy to help, a noted problem of
buddy systems (Cooper & Wight, 2014). However, if staff have received training
(Recommendation 1) this will decrease the frequency of communication required.

31
 Additionally, if staff experience busy periods, notifying their buddy and providing a
colleague’s email for urgent situations would resolve this. Oppositions may also arise
from staff unwilling to be involved, therefore, participation should be optional rather
than compulsory.

3.4 Action Plan

An action plan has been created (table5) to provide Organisation X a comprehensive

implementation guide detailing the key implementers and approach of each
recommendation. The plan also includes a timescale of when recommendations
could be implemented, and benefits will be experienced.

Table 5 Action Plan

32
.
Recommendation Rationale Timescale Key Organisational Implementation Cost/Resource Challenges Approaches to
(In priority order) Implementers benefits Approach requirements anticipated overcome
challenaes
1.Training Auditors are First steps InternalAudi Will reduce 1. Assess level of Low cost if Objections may Emphasise I
unable to of approach t People anxiety training required internaltrainin occur due to organisational
This use Al as to assess Manager (IA towards Al, through g cost of benefits of
theylack Al skills should improving evaluating current external using Al to
recommendaio HR) auditor skills
t n involves skills/ begin perceptions IA HR staff training improve
&
internal auditors knowl straight and determine and their time efficiency/accuracy
being Al trained edge and away acceptance expertisein which can reduce
haven't also organisatio Time of staff cost leading to
received Benefits n who will be cost savings
training, should be felt Being able trai ning & also
which also afterfirst to use Al will 2. If internal staff being
caused training increase training is trained
anxieties sessions efficiency required,create
in have been and training action
plan of skills that Between £950
participants completed enable auditors need and £2,000
to identify developed,by who per employee
additional and when depending on
risks, which session using
will improve If external an external
accuracy trainingis required, provider
contact external
training providers to
book training.
2.Meeting Participant Held as Internal Audit Benefci ial 1. Managemetn Low Objections by Explain howaim is
s soon as managemen in discussappropriat financial managers who not to resolve
This expressed possible and t reducingAl e worry they'II concerns, rather
related date/time cost
recommendation significant the benefits be unableto reduce/understand
uncertaintiesan for
involves a team anxiety will meeting Employeetime answer them
d anxieties
meeitng of and dislike likely follow which will Time is main questions/solv
Internal Auditors towardsAl immediatel improve 2. Email staff resource e concerns
about Al due to y after percetpionsan with details required for the
uncertainty d increase and a meeting and to
and If there is a acceptance of meeting summarise/typ Key points from
concerns development Al invite/Zoo e up key points People unable meeting
over how e.g., is set to m link to attend the distributed so
Al will be Benefci ial in meet ing due to those unable to
impact implemente termsof avoiding 3.Meeting availability
employee takes place attend can access
their role d this can be and is when convenient
demoivtation
1
3.5 Future Research

Through addressing the gap previously neglected in IA literature (Christ, Eulerich,

Krane, & Wood, 2020), this study contributes an original understanding of how AI
impacts the role of internal auditors. Adopting qualitative methods also enabled new
insights into how internal auditors perceive and feel towards AI. As global adoption of
AI is rapidly expanding (Singh, 2019), IA needs to prepare as failure to adapt will
diminish IAs ability to fulfil its purpose (IIA, 2020), as discussed in Report 1.
Therefore, these findings are relevant to organisations seeking to understand and
mitigate the impacts of AI through improving internal auditors’ adaptability and
acceptance. Moreover, this Enquiry illustrates the necessity of upskilling internal
auditors in AI, therefore integrating AI modules into accounting qualifications is a
policy recommended to professional accounting bodies. Nevertheless, applicability of
these findings may be restricted due to organisational context, as this Enquiry
focused on one organisation. As research in this area is limited, prospective efforts
could focus on the impact across other entities to reinforce these findings. Future
research could also explore accounting education at university level, to uncover
whether a curriculum change is necessary to prepare internal auditors for AI before
employment.
References
Abhabi, E., & Anozie, C. B. (2017). Literature Review for the Type of Interview in Qualitative
Research. International Journal of Education, 9(3), 1-12.
doi:https://doi.org/10.5296/ije.v9i3.11483

Afroze, D., & Aulad, A. (2020). Perception of Professional Accountants about the Application of
Artifical Intelligence (AI) in Auditing Industry of Bangladesh. Journal of Social Economics
Research, 7(2), 51-61. doi:10.18488/journal.35.2020.72.51.61

Ahmad, S., Wasim, S., Gogoi, S., & Ifran, S. (2019). Qualitative v/s Quantitative Research. Journal of
Evidence Based Medicine and Healthcare, 6(43), 2828-2832. doi:10.18410/jebmh/2019/587

Alshenqeeti, H. (2014). Interviewing as a Data Collection Method: A Critical Review. English

Linguistics Research, 3(1), 39-45. doi:https://doi.org/10.5430/elr.v3n1p39

Archibald, M. M., Ambagtsheer, R. C., Casey, M. G., & Lawless, M. (2019). Using Zoom
Videoconferencing for Qualitative Data Collection: Perceptions and Experiences of
Researchers and Participants. International Journal of Qualitative Methods, 18, 1-8.
doi:https://doi.org/10.1177/1609406919874596

Atieno, O. P. (2009). An Analysis of the Strengths and Limitation of Qualitative and Quantitative
Research Paradigms. Problems of Education in 21st Century, 13, 13-18.

Berger, R. (2015). Now I see it, now I don’t: researcher’s position and reflexivity in qualitative
research. Qualitative Research, 15(2), 219–234.
doi:https://doi.org/10.1177/1468794112468475

Bloomberg, L. D., & Volpe, M. (2019). Completing your Qualitative Dissertation: A Road Map from
Beginning to End (4th ed.). Los Angeles: Sage Publications LTD.

Bolderston, A. (2012). Conducting a Research Interview. Journal of Medical Imaging and Radiation
Sciences, 43, 66-76. doi:10.1016/j.jmir.2011.12.002

Braun, V., & Clarke, V. (2006). Using Thematic Analysis in Psychology. Qualitative Research in
Psychology, 3(2), 77-101. doi:10.1191/1478088706qp063oa

Bryman, A., & Bell, E. (2015). Business Research Methods (4th ed.).

Cassell, C. (2005). Creating the interviewer: identity work in the management research process.
Qualitative Research, 5(2), 167-179. doi:https://doi.org/10.1177/1468794105050833

Castleberry, A., & Nolen, A. (2018). Thematic analysis of qualitative research data: Is it as easy as it
sounds? Currents in Pharmacy Teaching and Learning, 10(6), 807-815.
doi:https://doi.org/10.1016/j.cptl.2018.03.019

Choy, L. T. (2014). The Strengths and Weaknesses of Research Methodology: Comparison and
Complimentary between Qualitative and Quantiative Approaches. IOSR Journal of
Humanities and Social Science, 19(4), 99-104. doi:10.9790/0837-194399104

Christ, M. H., Eulerich, M., Krane, R., & Wood, D. A. (2020). New Frontiers for Internal Audit
Research. doi: http://dx.doi.org/10.2139/ssrn.3622148

Clarke , V., & Braun, V. (2017). Thematic Analysis. The Journal of Positive Psychology, 12(3), 297-298.
doi:10.1080/17439760.2016.1262613

1
Connelly, L. M. (2016). Trustworthiness in Qualitative Research. Medsurg Nursing, 25(6), 435-436.

Cooper , J., & Wight, J. (2014). Implementing a buddy system in the workplace. Resource
Management . North America, Pheonix: Project Management Institute .

Cope, M. (2010). Coding Transcripts and Diaries. In N. Clifford, S. French, & G. Valentine , Key
Methods in Geography (pp. 440-450). SAGE.

Creswell, J. W. (2014). Research Design: Qualitative, Quantitative, and Mixed Method Approaches
(4th ed.). Thousand Oaks: SAGE Publications.

Darino, L., Hancock, B., & Lazaroff-Puck, K. (2019, December 9). Employee Motivation in the Age of
Automation and Agility. Retrieved from Mckinsey Institute:
https://www.mckinsey.com/business-functions/organization/our-insights/the-organization-
blog/employee-motivation-in-the-age-of-automation-and-agility#

Davis, F. D. (1989). Perceived Usefulness, Perceived Ease of Use, and User Acceptance. MIS
Quarterly, 319-340. doi:https://doi.org/10.2307/249008

Deloitte. (2020). The Upskilling Imperative: Building a future-ready workforce for the AI age.
Retrieved from Deloitte:
https://www2.deloitte.com/content/dam/Deloitte/ca/Documents/deloitte-analytics/ca-
covid19-upskilling-EN-AODA.pdf

DiCicco-Bloom, B., & Crabtree, B. F. (2006). The qualitative research interview. Medical Education,
40(4), 314-321. doi:https://doi.org/10.1111/j.1365-2929.2006.02418.x

Fisher, A. (2019, November 9). Workers Are Worried Robots Will Steal Their Jobs. Here’s How to Calm
Their Fears. Retrieved from Fortune: https://fortune.com/2019/11/09/workplace-employee-
fears-robots-artificial-intelligence/

Forbes. (2019, May 22). How to Prep Your Employees for AI Disruption. Retrieved from Forbes:
https://www.forbes.com/sites/insights-intelai/2019/05/22/how-to-prep-your-employees-
for-ai-disruption/?sh=76a9967f6204

Fox, N. (2009). Using Interviews in Research Project. The NIHR RDS for the East Midlands/Yorkshire,
1-37. Retrieved from https://www.rds-yh.nihr.ac.uk/wp-
content/uploads/2013/05/15_Using-Interviews-2009.pdf

Galletta, A. (2013). Mastering the Semi-Structured Interview and Beyond: From Research Design to
Analysis and Publication. New York: NYU Press.

Gardener, M. (2020, March 20). How to Enourage Informal Learning in your Organization. Retrieved
from Forbes: https://www.forbes.com/sites/forbesbusinesscouncil/2020/03/20/how-to-
encourage-informal-learning-in-your-organization/?sh=4e011ce0526b

Gray, L. M., Wong-Wylie, G., Rempel, G. R., & Cook, K. (Eds.). (2020). Expanding Qualitative Research
Interviewing Strategies: Zoom Video Communications. The Qualitative Report, 25(5), 1292-
1301.

Guest, G., Bunce, A., & Johnson, L. (2006). How Many Interviews Are Enough?: An Experiment with
Data Saturation and Variability. Field Methods, 18(1), 59-82.
doi:10.1177/1525822X05279903

2
Hammarberg, K., Kirkman, M., & Lacey, S. d. (2016). Qualitative Research Methods: When to use
them and how to judge them. Human Reproduction, 31(3), 498-501.
doi:doi:10.1093/humrep/dev334

Haynes, K. (2012). Reflexivity in Qualitative Research. In G. Symon, & C. Cassell, Qualitative

Organizational Research: Core Methods and Current Challenges (pp. 72-89). SAGE
Publications . doi:http://dx.doi.org/10.4135/9781526435620.n5

Hofisi, C., Hofisi, M., & Mago, S. (2014). Critiquing Interviewing as a Data Collection Method.
Mediterranean Journal of Social Sciences, 5(16), 60-64. doi:10.5901/mjss.2014.v5n16p60

HR News. (2018, September 21). How can external staff training benefit your business? Retrieved
from HR News: http://hrnews.co.uk/how-can-external-staff-training-benefit-your-business/

IIA. (2020). Future-proofing internal audit. Chartered Institute of Internal Auditors. Retrieved from
file:///C:/Users/W17036822/Downloads/Future-proofing%20internal%20audit%20(2).pdf

Jacobson, D., & Mustafa, N. (2019). Social Identity Map: A Reflexivity Tool for Practicing Explicit
Positionality in Critical Organizational Research. International Journal of Qualitative
Methods, 18, 1-12. doi:https://doi.org/10.1177/1609406919870075

Jamshed, S. (2014). Qualitative research method-interviewing and observation. Journal of Basic and
Clinical Pharmacy, 5(4), 87-88. doi:10.4103/0976-0105.141942

Janghorban, R., Roudsari, R. L., & Taghipour, A. (2014). Skype interviewing: The new generation of
online synchronous interview in qualitative research. International Journal of Qualitative
Studies on Health and Well-being, 9(1), 1-3. doi:https://doi.org/10.3402/qhw.v9.24152

Javadi, M., & Zarea, K. (2016). Understanding Thematic Analysis and its Pitfall. Journal of Client Care,
1(1), 34-40. doi:10.15412/J.JCC.02010107

JBI Training. (2021). AI & Data Science. Retrieved from JBI International :
https://jbinternational.co.uk/page/66/ai

Kaiser, K. (2009). Protecting Respondent Confidentiality in Qualitative Research. Qualitative Health

Research, 19(11), 1632-1641. doi:10.1177/1049732309350879

Kim, H.-J., Mannino, M., & Nieschwietz, R. J. (2009). Information technology acceptance in the
internal auditprofession: Impact of technology features and complexity. Internal Journal of
Accounting Information Systems, 10(4), 214-228.
doi:https://doi.org/10.1016/j.accinf.2009.09.001

Knox, S., & Burkard, A. W. (2008). Qualitative Research Interviews. Psychotherapy Research, 19(4-5),
566-575. doi:10.1080/10503300802702105

Kotb, A., Elbardan, H., & Halabi, H. (2020). Mappinh of Internal Audit Research: A post-Enron
structured literature review. Accounting, Auditing and Accountability Journal, 33(8), 1969-
1996. doi:10.1108/AAAJ-07-2018-3581

Lincoln, Y. S., & Guba, E. (1985). Naturalistic Inquiry. International Journal of Intercultural Relations,
9(4), 438-439. doi:10.1016/0147-1767(85)90062-8

Lolacono, V., Symonds, P., & Brown, D. H. (2016). Skype as a Tool for Qualitative Research
Interviews. Sociology Research Online, 21(2), 103-117. doi:https://doi.org/10.5153/sro.3952

3
Maguire, M., & Delahunt, B. (2017). Doing a Thematic Analysis: A Practical, Step-by-Step Guide for
Learning and Teaching Scholars. All Ireland Journal of Teaching and Learning in Higher
Education, 8(3), 3351-33514. Retrieved from http://ojs.aishe.org/index.php/aishe-
j/article/view/335

Manderson, L., Bennett, E., & Andajani-Sutjahjo, S. (2006). The Social Dynamics of the Interview:
Age, Class, and Gender. Qualitative Health Research, 16(10), 1317-1334.
doi:10.1177/1049732306294512

McCusker, K., & Gunaydin, S. (2014). Research using Qualitative Quantitative or Mixed Methods
Choice Based on the Research. Perfusion, 30(7), 537–542.
doi:https://doi.org/10.1177/0267659114559116

McFayden, J., & Rankin, J. (2016). The Role of Gatekeepers in Research: Learning from Reflexivity
and Reflection. GSTF Journal of Nursing and Health Care, 4(1), 82-88. doi:10.5176/2345-
718X_4.1.135

McKendrick, J. (2017, January 24). Artificial Intelligence Doesn't Just Cut Costs, It Expands Business
Brainpower. Retrieved from Forbes:
https://www.forbes.com/sites/joemckendrick/2017/01/24/artificial-intelligence-doesnt-
just-cut-costs-it-expands-business-brainpower/?sh=2c1875b5535a

Mohajan, H. (2018). Qualitative research methodology in social sciences and related subjects .
Journal of Economic Development, Environmental and People, 7(1), 23-48.
doi:10.26458/jedep.v7i1.571

Moon, K., Brewer, T. D., Hartley, S., Adams, V. M., & Blackman, D. A. (2016). A Guide to Improve
Qualitative Social Science Publishing in Ecology and Conservation Journals. Ecology and
Society, 21(3). doi:10.5751/ES-08663-210317

Nijhawan, L. P., Janodia, M. D., Muddukrishna, B. S., Bhat, K. M., Bairy, K. L., Udupa, N., & Musmade,
P. B. (2013). Informed consent: Issues and challenges. Journal of Advanced Pharmaceutical
Technology & Research, 4(3), 134-140. doi:10.4103/2231-4040.116779

Nowell, L. S., Norris, J. M., White, D. E., & Moules, N. J. (2017). Thematic Analysis: Striving to Meet
the Trustworthiness Criteria. International Journal of Qualitative Methods, 16, 1-13.
doi:https://doi.org/10.1177/1609406917733847

Patton, M. Q. (2002). Qualitative research and evaluation methods. Thousand Oaks: SAGE
Publications.

Querios , A., Faria, D., & Almeida, F. (2017). Strengths and Limitations of Qualitative and Quantitative
Research Methods. European Journal of Education Studies, 3(9), 369-387.
doi:10.5281/zenodo.88708

Roulston, K., & Choi, M. (2017). Qualitative Interviews. In U. Flick, The SAGE Handbook of Qualitative
Data Collection (pp. 233-246). SAGE.

Saunders, M. N., Thornhill, A., & Lewis, P. (2017). Research Methods for Business Students (8th ed.).
Harlow: Pearson.

Singh, P. (2019, January 22). Enterprise Use of AI has Grown 270 per cent Globally Over the Past Four
Years. Retrieved from Entrepreneur: https://www.entrepreneur.com/article/326737

4
Staller, K. M. (2015). Qualitative Analysis: The art of building bridging relationships. Qualitative Social
Work: QSQ: Research and Practice, 14(2), 145-153. doi:10.1177/1473325015571210

Tai, M. C.-T. (2012). Deception and informed consent in social, behavioral, and educational research
(SBER). Tzu Chi Medical Journal, 24(4), 218-222. doi:10.1016/j.tcmj.2012.05.003

Tessier, S. (2012). From Field Notes, to Transcripts, to Tape Recordings: Evolution or Combination?
International Journal of Qualitative Methods, 11(4), 446-460.
doi:https://doi.org/10.1177/160940691201100410

Venkatesh, V., & Davis, F. D. (2000). A Theoretical Extension of the Technology Acceptance Model:
Four Longitudinal Field Studies. Management Science, 2, 186-204.
doi:10.1287/mnsc.46.2.186.11926

Wa-Mbaleka, S. (2019). The Researcher as an Instrument. In A. P. Costa, L. P. Reis, & A. Moreira,

Computer Supported Qualitative Research. (Vol. 1068, pp. 33-41). Springer: Springer.
doi:https://doi.org/10.1007/978-3-030-31787-4_3

Weller, S. (2015). The potentials and pitfalls of using Skype for qualitative (longitudinal) interviews.
National Centre for Research Merhods Working Paper, 1-49.

Whitman, C., & Sobczak, M. (2018). AI: Overrated or the Future of Accounting. Undergraduate
Student Research Awards, 45, 1-26.

Zakaria, B. R., & Mustamal, A. H. (2014). Rapport Building in Qualitative Research. 1-9. Retrieved
from
http://eprints.utm.my/id/eprint/61304/1/AedeHatibMustaamal2014_RapportBuildinginQua
litativeResearch.pdf

5
 Appendices:

Appendix A: Reflective Statement

Figure 2 Undergraduate Programme Goals

Throughout this Enquiry, I have developed several skills that have facilitated
personal development and enabled me to achieve the specific objectives highlighted
above, Using the feedback that I received on my Enquiry proposal allowed me to
refine the focus of my topic to explore artificial intelligence, rather than automation
as originally proposed. This directed the focus of my literature review; however, I
encountered a problem here as I had never written critically or conducted a review
before, meaning I felt unsure what to do. I dealt with this through engaging with
online critical writing resources and reading literature reviews. This developed my
research and critical thinking skills (2.2) and meant I was able to challenge the
thoughts of scholars throughout my review (2.2). This identified several gaps in
knowledge that I later addressed in my study, enabling me to conduct innovative
research (1.3) to make an original contribution to the topic. Through the review, I
gained an understanding of theory relevant to the topic, which I utilised to assess
professional practice at Organisation X (1.1) and identify recommendations to
alleviate problems encountered related to AI (1.2). Upon reflection, a problem I faced
throughout the Enquiry related to using time effectively, as I often fell behind
schedule, meaning I would also become overwhelmed. This problem mainly

6
stemmed from the personal struggles I have with self-doubt, as due to this, I spent
time too much time worrying if my ideas were “correct” or good enough, which
prevented me from working and meant I required reassurance often. However, I feel
as though through undertaking the Enquiry and the support I have received from my
supervisor, my self-confidence has developed. As result, I have also learned to
believe in myself more, which I will take forward into my professional career. If I was
undertaking the Enquiry again, I would consider exploring the issue in an
international context, as this would enable me to achieve objective 1.4. In terms of
future learning needs discerned from this reflection, I do think my time management
(3.1) still requires improvement. Therefore, I plan to develop this through setting
myself mini deadlines to ensure I am on schedule, as good time management will be
essential when completing projects throughout my career.

7
Appendix B: Interview Schedule
Introductory questions:
How many years have you worked as an internal auditor?
What department do you audit?
1.Role:
In your opinion, what is the role of the Internal Audit in the organisation?
How do you believe this role will be impacted in future as a result of AI?
How do you believe your current job role will be impacted by AI?
Research says role in future will be ‘hybrid’ of data analyst and internal auditor, what do you
think about this?
2.AI:
Are there any challenges you anticipate in relation to the use of AI?
Do you have any apprehensions about using AI? If so, what are they?
3.Technology Acceptance Model:
Perceived Usefulness
How do you believe the use of AI would be beneficial in your role?
To what extent do you believe using AI is relevant to your role?
Perceived Ease-of-Use
Do you think AI would be easy to use in your role?
How do you think the skills required in your role will change as more AI is introduced/role
evolves towards data analytics?

8
Appendix C: Phase 3 Thematic Analysis Coding

9
Appendix D: Organisation Informed Consent

10
Appendix E: Informed Consent

11
Appendix F: Participant Brief

12
13