2024 Proceedings of the ISCAP Conference ISSN: 2473-4901

Baltimore, MD v10 n6119

Navigating the AI Landscape: An Analysis of AI

Developer Tools Usage, Sentiment, and Trust
Among IT Professionals

Wendy Ceccucci
[email protected]
Business Analytics and Information Systems Department
Quinnipiac University
Hamden, CT 06518 USA

Alan Peslak
[email protected]
Department of Information Sciences and Technology
Penn State University
Dunmore, PA 18512 USA

Kiku Jones
[email protected]
Business Analytics and Information Systems Department
Quinnipiac University
Hamden, CT 06518 USA

Abstract
GitHub Copilot, developed by GitHub, is a new tool aiding developers with a range of tasks, including
the generation of code snippets, documentation assistance, and the formulation of implementation
strategies. Comparable AI development tools, such as Tabnine and AWS Code Whisperer, also serve as
developmental aids but are utilized to varying and much lesser extents. Our research examines the
adoption of GitHub Copilot and similar AI development tools among professional developers and other
users using the Stack Overflow Annual Survey. The study reveals notable age-related disparities in tool
usage, with younger individuals showing a markedly higher propensity to employ these technologies
compared to older users. Additional significant insights include variations in usage based on the type of
developer, professional status, and the influence of user attitudes towards AI on the adoption of GitHub
Copilot among developers.

Keywords: AI Developer Tools, Artificial Intelligence, GitHub Copilot, Technology Adoption, AI Trust

1. INTRODUCTION automated code assistants to advanced

debugging algorithms, have become pivotal in
The integration of Artificial Intelligence (AI) in shaping how IT professionals approach software
software development has marked a development. This paper presents a
transformative era in the information technology comprehensive analysis of the usage patterns,
(IT) sector. AI developer tools, ranging from sentiment, and trust levels associated with AI

©2024 ISCAP (Information Systems and Computing Academic Professionals) Page 1

https://iscap.us/proceedings/
2024 Proceedings of the ISCAP Conference ISSN: 2473-4901
Baltimore, MD v10 n6119

developer tools among IT developers. The study these aspects, the study seeks to contribute to
aims to provide an understanding of how these the responsible and effective integration of AI in
tools are reshaping the development landscape, software development practices.
the attitudes of developers towards them, and the
trust issues that may arise. 2. LITERATURE REVIEW

The advent of AI in development tools has AI Code Generation Tools

introduced both opportunities and challenges. Ai code generation systems are simply AI systems
While these tools promise increased efficiency, trained to write code, thereby automating the
accuracy, and even creativity in coding, they also process of programming. The code generating
raise questions about reliability, ethical use, and system is built upon machine learning algorithms
the potential for diminishing human skill sets. that train on large datasets of code. The
Understanding the sentiments and trust levels of algorithms learn the common patterns, practices,
developers towards AI tools is crucial for the and conventions in coding and then can generate
future design, implementation, and governance new code based on their learning.
of these technologies. According to a developer
survey by GitHub (2024) 92% of developers AI coding tools are becoming standard practice
working in large companies are already using AI for many software developers. According to an
tools either at work or in their personal time. online survey conducted by Wakefield Research
on the behalf of GitHub, 92% of U.S. developers
This paper begins by exploring the current are using AI coding tools both in and outside of
landscape of AI developer tools, categorizing work (Shani & GitHub Staff, 2023). The survey
them based on their functionalities and the responses were from 500 U.S. based developers
aspects of software development they impact. We who were not managers and worked at
then delve into the methodologies used to gauge companies with 1,000-plus employees. They also
the usage patterns of these tools among IT found that more than 4 out of 5 developers
professionals. This includes a survey of a diverse expected AI coding tools would make their team
group of developers, encompassing various more collaborative.
industries, experience levels, and geographical
locations. The most widely used AI developer tool is
GitHub’s Copilot which was released for technical
Following the usage analysis, the paper addresses preview in 2021. GitHub Copilot is an AI-powered
the core aspects of sentiment and trust. code assistant developed by GitHub in
Sentiment analysis is conducted through survey collaboration with OpenAI. GitHub Copilot is
responses. This analysis provides insights into the designed to help developers write better code
general attitudes of developers towards these faster and with fewer errors. It operates by
tools, ranging from enthusiasm and optimism to analyzing the context within the codebase and
skepticism and concern. Trust, a more complex suggesting whole lines or blocks of code that
and multifaceted issue, is examined through a developers can incorporate or modify as needed.
separate survey response.
GitHub Copilot, an AI-powered code assistant, is
The findings of this study are expected to offer used by software developers primarily in
valuable insights for multiple stakeholders. Tool JavaScript and Python, with Visual Studio Code
developers and AI researchers can gain a better being the main IDE. It is most commonly used for
understanding of user attitudes and trust factors, data processing and code generation, with the
guiding them in creating more user-centric and potential to significantly reduce development
trustworthy tools. Organizations and team time and slightly increase code quality. However,
leaders can use these insights to make informed concerns about security and difficulty of
decisions about integrating AI tools into their integration have been raised, and developers are
workflows, considering both the technical and divided in their opinions about the tool (Jaworski
human aspects. Additionally, the study & Piotrkowski, 2023; Zhang, et al., 2023).
contributes to the broader discourse on the role
of AI in the future of work, particularly in the IT Other examples of AI based code assistants
sector. include Tabnine (Kedar, 2024), and
CodeWhisperer (Desai & Deo 2022). Comparing
This paper aims to shed light on the complex the tools:
dynamics between IT developers and AI 1. GitHub Copilot is owned by Microsoft and
developer tools, focusing on usage patterns, is based on OpenAI’s Codex model. One
sentiments, and trust. By providing an analysis of of its strengths is its flexibility in

©2024 ISCAP (Information Systems and Computing Academic Professionals) Page 2

https://iscap.us/proceedings/
2024 Proceedings of the ISCAP Conference ISSN: 2473-4901
Baltimore, MD v10 n6119

deployment (cloud vs local). It excels in Reussner, 2006; Lipner, 2004).

its deep integration with GitHub and its 2. Contextual understanding. AI tools often
powerful language model but is best fail to fully grasp specific project
suited for those already invested in the contexts, resulting in misaligned code
GitHub ecosystem. (Wang, et al., 2024).
2. Tabnine based on various models 3. Dependency on data. The effectiveness of
including GPt-3. It also offers flexibility in AI code generation tools is highly
deployment and a wide range of IDE and dependent on the quality and breadth of
language support, which is beneficial for their training data. Poor or biased data
diverse development environments. leads to poor performance.
3. Amazon CodeWhisperer is developed by 4. Security concerns. AI-generated code
Amazon Web Services (AWS). It is based may introduce vulnerabilities if secure
on AWS’s proprietary models, including coding practices are not properly
adaptations of large language models. embedded. According to a study by Snyk
The context-aware code suggestions are (2023), more than three-quarters of
similar to the other tools but tailored for developers bypass established protocols
cloud and serverless environments. to use code completion tools despite
potential risks.
According to a Survey by CodeSignal (2024) of 5. Limited creativity. AI lacks the creativity
1,000 developers worldwide, 81% of developers and problem-solving abilities inherent to
surveyed said they use AI-powered coding human developers, leading to less
assistants. The number one reason for using the innovative solutions.
tool was for learning new technical skills or 6. Trust and reliability. Developers often find
knowledge. The second most common reason it challenging to trust AI-generated code
was for generating boilerplate code. without extensive testing and
verification, increasing their workload
Some of the advantages of AI code generation (Wang, et al., 2024).
tools include: 7. Ethical and legal issues. The use of AI in
1. Saves time. These tools can accelerate code generation raises concerns related
the development cycle, making the to intellectual property and the legality of
process more efficient. using certain code snippets.
2. Learning and skill development. AI code
generation supports learning and skill According to a study by Dakhel, et al. (2023),
enhancement. By observing the AI- Copilot can be an asset in software projects when
generated code developers can learn new used by expert developers, as its suggestions can
techniques, understand different code match human contributions in quality. However,
structures, and adopt better coding Copilot can become a liability if used by novice
practices. developers who may struggle to filter its
3. Accessibility for non-experts. Individuals suggestions effectively.
with less coding knowledge can create
functional applications by simply Trust and AI Tools
describing the requirements in natural Recent studies have explored the impact of AI-
language. powered code generation tools on software
4. Reduces human error. AI tools can help developers' trust and the potential for these tools
minimize human errors and the AI- to automate routine programming tasks (Cheng
generated code is often well-documented et al., 2024; Ernst & Bavota, 2022). These tools,
with explanations making it easier to such as GitHub Copilot, have the potential to
understand and debug. increase the level of abstraction in software
development, allowing developers to focus on
Along with its potential advantages, AI code business processes rather than code (Palacios-
generated tools have some limitations. These González, et al., 2008). However, concerns have
include: been raised about the potential for bias, legal
compliance, and security vulnerabilities in AI-
1. Quality and accuracy. AI-generated driven development environments (Ernst &
code may not always meet the quality Bavota, 2022).
and accuracy standards required for
complex projects (Hasselbring & One significant challenge involves helping users
evaluate the trustworthiness of AI tools. The
importance of software developers' trust in

©2024 ISCAP (Information Systems and Computing Academic Professionals) Page 3

https://iscap.us/proceedings/
2024 Proceedings of the ISCAP Conference ISSN: 2473-4901
Baltimore, MD v10 n6119

programming has been extensively studied, learning courses into their curricula, which equips
highlighting it as a crucial design requirement for new graduates with the skills and familiarity
these tools. Trust is considered a key prerequisite needed to leverage these tools effectively.
for ensuring the safety of the resulting software
products (Hasselbring & Reussner, 2006; Lipner, Moreover, the rapid pace of technology means
2004). that younger individuals often have a 'digital
native' advantage. They tend to be more adaptive
According to a study by Wang, et al. (2024), to changes in the tech landscape, including the
developers’ trust is rooted in the AI tool’s use of sophisticated AI platforms that require a
perceived ability, integrity, and benevolence, and steep learning curve. A report by GitHub, The
is situational, varying according to the context of State of the Octoverse (Daigle & GitHub Staff,
usage. Existing AI code generation tools lack the 2024), highlights that newer programmers are
affordances for developers to efficiently and more likely to utilize AI coding assistants,
effectively evaluate the trustworthiness of AI- attributing this to their up-to-date training and
powered code generation tools. inherent flexibility in adopting new workflows.

Several other demographics and factors can Additionally, the software development industry
influence a developer's trust in AI tools, and itself fosters a culture of innovation and
understanding these can help tailor AI solutions continuous learning, which resonates well with
to better meet user needs. These factors include the mindset of younger developers. They are
developer’s geographical location, gender, and often driven by the potential to streamline coding
socioeconomic status. processes, enhance productivity, and solve
complex problems efficiently with the help of AI
Developers from different regions may have tools. In practical terms, younger developers
varying levels of trust in AI tools based on cultural utilize AI tools for a range of tasks including
attitudes towards technology and data privacy. writing and reviewing code, automating repetitive
Although their research did not focus on tasks, and even in conceptual phases of
developers specifically, Grassini and Ree (2023) development like brainstorming and prototyping.
found that respondents from the USA The convenience and efficiency offered by AI tools
demonstrated higher levels of hopefulness for AI align with the fast-paced, results-oriented
technology compared to those from the UK. Their environment preferred by this demographic. The
finding suggests that cultural context does play a influence of AI on younger developers is not just
role in shaping individuals' perceptions of AI transforming their individual workflows but is also
technology. shaping the future dynamics of team
collaborations and project management in
In the same study, they found that male software development. As these young
respondents showed higher hopes for AI systems developers progress in their careers, their
than female respondents. This finding is preference for AI-enhanced workflows is likely to
consistent with previous research, which reported catalyze broader adoption of these technologies
that males perceive AI as more useful (Arujo, et across the industry, potentially leading to more
al., 2020) and generally more favorable than innovative solutions in the overall approach to
females (Lozano, et al., 2021). Likewise, a study software development.
by Armutat, et al. (2024) found that men tended
to view AI applications more positively, rate their 3. METHODOLOGY
own AI competencies higher, and have more trust
in the technology compared to women. This study explores the usage of AI tools by
software developers by leveraging data from the
Age and AI Coding Tools 2023 Stack Overflow survey. The Stack Overflow
The integration of AI tools into the workflow of Developer Survey is widely acknowledged as the
software developers has shown a notable age- most comprehensive survey targeting coders
related trend, with younger developers worldwide. Annually, it covers a wide range of
demonstrating a higher propensity to adopt such subjects, from developers’ technology
technologies. This trend can be attributed to preferences to their career aspirations. According
several factors. First, younger developers are to Stack Overflow’s website:
generally more exposed to the latest “For 13 years, we've delivered industry-
technological innovations during their education leading insights regarding the developer
and early careers, making them more receptive community. This is the voice of the
to AI-driven solutions. Educational institutions developer. Analysts, IT leaders,
are increasingly incorporating AI and machine reporters, and other developers turn to

©2024 ISCAP (Information Systems and Computing Academic Professionals) Page 4

https://iscap.us/proceedings/
2024 Proceedings of the ISCAP Conference ISSN: 2473-4901
Baltimore, MD v10 n6119

this report to stay up to date with the

evolving developer experience,
technologies that are rising or falling in
favor, and to understand where tech
might be going next.” (Stack Overflow,
2023)

The validity of using Stack Overflow data is

supported by its frequent citation in numerous
peer-reviewed publications, including studies by
Barua, et al. (2014), Asaduzzaman, et al. (2013),
and Treude and Robillard (2016). The dataset Table 2: AI- Powered Developer Tool Usage
by Software Developers
comprises a rich mix of demographic data,
descriptive statistics, and responses to opinion-
The main AI developer tool is clearly GitHub
based questions about the programming
Copilot, with approximately 25% usage by both
industry. IBM SPSS 29 was utilized for the data
analysis. Primarily descriptive statistics and groups. The next most used tool is Tabnine, at
crosstab analyses were used in the study. about 5%. Given the dominance of GitHub
Copilot, the focus of the rest of our paper will be
on this tool.
4. RESULTS
The next question related to AI development tools
Survey respondents were asked, "Which AI-
usage asked the question: “Do you currently use
powered developer tools did you use regularly
over the past year and which do you plan to work AI tools in your development process?” The three
with over the next year?" possible responses were “Yes”, “No, but I plan to
soon”, and “No, I don’t plan to”. Table 3 shows
the results. Overall, the study reveals that 44%
Table 1 displays the results of this survey
of Developers currently use AI tools and another
question for all participants. Out of 89,870 total
26% plan to soon. Only 30% do not have plans
responses, 686 respondents did not answer the
to use AI.
question. This leaves a total of 89,184 valid
responses.

Table 3: Plans to Use AI- Powered Developer

Tool by Software Developers

The survey then looked at how respondents

Table 1: AI- Powered Developer Tool Usage viewed AI (Table 4). Overall, 76% of developers
by All Respondents have a favorable view of AI. Table 5 shows the
effect of attitude towards AI and the use of GitHub
In the next table, Table 2, the responses were Copilot. Those with a very favorable view use
limited to software developers only. A total of Copilot nearly 50%. This rate drops for other
67,973 respondents identified as software views hovering in the 20% range for those with
developers by profession. With 686 respondents neutral or unfavorable views.
not answering the question, the total number of
valid responses was 67,287.

©2024 ISCAP (Information Systems and Computing Academic Professionals) Page 5

https://iscap.us/proceedings/
2024 Proceedings of the ISCAP Conference ISSN: 2473-4901
Baltimore, MD v10 n6119

somewhat trust AI output using GitHub Copilot.

But the percentage of developers with neutral or
lack of trust only drops to about 32%. This
finding, though significant, demonstrates that
trust has a limited impact on the use of AI
development tools.

Table 4: Software Developer’s Attitudes

Toward AI

Table 7: Software Developer’s Trust in AI

and use of GitHub Copilot

Age is clearly a factor in the use of GitHub Copilot

(Table 8). This result has been supported by the
literature. Nearly one half of the developers
under 18 use it and the percentage declines for
each older age group. The 18-24 users check in
at 35%, 25-34 at 27% and so on. The oldest
Table 5: Software Developer’s Attitudes group over 65 only has a 6% participation rate.
Toward AI and use of GitHub Copilot

The survey then asked respondents “How much

do you trust the accuracy of the output from AI
tools as part of your development workflow?”
Table 6 shows the survey results. Sentiment is
higher than trust in the accuracy of AI output with
only 2% of developers having high trust. There is
a large percentage who somewhat trust AI at
38%. Twenty-eight percent distrust AI and 32%
are neutral.

Table 8: Software Developer’s Age and use

of GitHub Copilot
Table 6: Software Developer’s Attitudes
Toward AI and use of GitHub Copilot When we examine all respondents, we find that
both hobbyists and developers by profession have
The effect of trust on the usage of GitHub Copilot the highest use of GitHub Copilot (Table 9).
is revealing. Lack of trust negatively affects
adoption with 39% of those who at least

©2024 ISCAP (Information Systems and Computing Academic Professionals) Page 6

https://iscap.us/proceedings/
2024 Proceedings of the ISCAP Conference ISSN: 2473-4901
Baltimore, MD v10 n6119

Surprisingly, those learning to code are lower Showing the benefits of learning to code utilizing
than these groups at 23%. the AI developer tools and the time saving in
completing projects will likely be a positive for
both the organization and the developer. If there
are developers on site who are already proficient
in using the AI developer tools, it may be
beneficial to the organization to have these
experienced developers work with those who are
just beginning to discover the AI developer tools.
In addition, those developer types who have the
highest usage rate of AI developer tools may be
the best people to talk to the other developers
who are not utilizing the tools to discuss their
experiences to provide credibility to the
recommendation to use the tools.

6. REFERENCES
Table 9: Developer’s Level and use of GitHub Araujo, T., Helberger, N., Kruikemeier, S. &
Copilot Vreese, C. (2020). In AI we trust?
Perceptions about automated decision-
When we examine the developer type and use of making by artificial intelligence. AI &
GitHub Copilot, we find that Blockchain, Society, 35, 611-623.
Developer Advocates, and Front-end developers https://doi.org/10.1007/s00146-019-
have the highest usage rate. They all exceed 00931-w
30% usage (Appendix 1). Database, enterprise,
and embedded developers all are lowest end at Armutat, S., Wattenber, M., & Mauritz, N. (2024).
less than 15%. Artificial Intelligence – Gender-Specific
Differences in Perception, Understanding,
5. CONCLUSIONS and Training Interest. Proceedings of the 7th
International Conference on Gender
GitHub Copilot was found to be used the most of Research, 7(1), 36-43.
all AI powered developer tools regardless of a https://doi.org/10.34190/icgr.7.1.2163
person’s profession. When looking strictly at Asaduzzaman, M., Mashiyat, A. S., Roy, C., &
software developers, the same result was found Schneider, K. (2013). Answering questions
with GitHub Copilot being used by 25% of the about unanswered questions of stack
respondents and the next tool coming in at only overflow. Proceedings of the 10th Working
5%. An additional finding to help tool developers Conference on Mining Software Repositories,
is that a majority of developers have a favorable 97-100.
view of AI and somewhat trust it. The favorable https://doi.org/10.1109/MSR.2013.662401
view of AI does seem to impact a developer’s 5
choice in using the AI developer tool. However,
trust in AI appears to have limited impact. This is Barua, A., Thomas, S., & Hassan, A. (2014). What
an area where future researchers will want to look are developers talking about? An analysis of
further to determine what other factors are topics and trends in stack
possibly moderating this relationship. overflow. Empirical Software
Engineering, 19, 619-654.
Age has a clear impact on the use of AI developer https://doi.org/10.1007/s10664-012-9231-
tools. As the age group increased, the use of AI y
developer tools decreased. This finding, which is
Cheng, R., Wang, R., Zimmermann, T., & Ford,
supported in the literature, may be helpful to
D. (2024). “It would work for me too”: How
organizations as they review their workforce and
Online Communities Shape Software
determine potential training or motivating
Developers’ Trust in AI-Powered Code
opportunities to engage different age groups to
Generation Tools. ACM Transactions on
participate in using the AI developer tools.
Interactive Intelligent Systems. 14(2), 1–39.
Another finding which can be helpful to
https://doi.org/10.1145/3651990
organizations is that those learning to code do not
use the AI developer tools as much as CodeSignal (2024). Trends Report Developers &
professional developers or even hobbyists. AI Coding Assistant Trends. CodeSignal.

©2024 ISCAP (Information Systems and Computing Academic Professionals) Page 7

https://iscap.us/proceedings/
2024 Proceedings of the ISCAP Conference ISSN: 2473-4901
Baltimore, MD v10 n6119

Retrieved July 26, 2024, from Lipner, S. (2004). The trustworthy computing
https://codesignal.com/report-developers- security development lifecycle. Proceedings
and-ai-coding-assistant-trends/ of the 20th Annual Computer Security
Applications Conference, 2–13.
Daigle, K. & GitHub Staff (2024). The State of the
https://doi.org/10.1109/CSAC.2004.41
Octoverse. GitHub. Retrieved on July 26,
2024, from https://GitHub.blog/news- Lozano, I., Molina, J. & Gijón, C. (2021).
insights/research/the-state-of-open-source- Perception of artificial intelligence in
and-ai/ Spain. Telematics and Informatics,
63. https://doi.org/10.1016/j.tele.2021.101
Desai, A. & Deo, A. (2022). Introducing Amazon
672
CodeWhisperer, the ML-powered coding
companion. AWS. Retrieved July 26, 2024, Palacios-Gonzalez, E., Fernandez-Fernandez, H.,
from Diaz, V., Garcia-Bustelo, B., & Lovelle, J.
https://aws.amazon.com/blogs/machine- (2008). A review of Intelligent Software
learning/introducing-amazon-codewhisperer- Development Tools. Proceedings of the 2008
the-ml-powered-coding-companion/ International Conference on Artificial
Intelligence, 585-590.
Dakhel, A., Majdinasab, V., Nikanjam, A., Khomh,
F., Desmarais, M., & Jiang, Z. (2023). GitHub Shani, I., & GitHub Staff (2023). Survey reveals
Copilot AI pair programmer: Asset or AI’s impact on the developer experience.
Liability?, Journal of Systems and Software, GitHub. Retrieved July 26, 2024, from
203. https://GitHub.blog/2023-06-13-survey-
https://doi.org/10.1016/j.jss.2023.111734 reveals-ais-impact-on-the-developer-
experience/
Ernst, N., & Bavota, G. (2022). AI-Driven
Development Is Here: Should You Worry?
Snyk (2023). 2023 Snyk AI-Generated Code
IEEE Software, 39(2) 106-110.
Security Report. Snyk. Retrieved on July 26,
https://doi.org/10.48550/arXiv.2204.07560
2024, from https://snyk.io/reports/ai-code-
GitHub (2024). GitHub Developer Survey. security/
GitHub. Retrieved July 26, 2024, from
Stack Overflow Retrieved July 26, from
https://GitHub.blog/news-
https://survey.stackoverflow.co/2020#over
insights/research/survey-reveals-ais-impact-
view
on-the-developer-experience/#developers-
want-more-opportunities-to-upskill-and- Treude, C., & Robillard, M. P. (2016). Augmenting
drive-impact API documentation with insights from stack
overflow. Proceedings of the 38th
Grassini, S. & Ree, A. (2023). Hope or Doom AI-
International Conference on Software
ttitude? Examining the Impact of Gender,
Engineering, 392-403.
Age, and Cultural Differences on the
https://doi.org/10.1145/2884781.2884800
Envisioned Future Impact of Artificial
Intelligence on Humankind. Proceedings of Wang, R., Cheng, R., Ford, D., & Zimmerman, T.
the European Conference on Cognitive (2024). Investigating and Designing for
Ergonomics 2023, 1-7. Trust in AI-powered Code Generation Tools.
https://doi.org/10.1145/3605655.3605669 Proceedings of the 2024 ACM Conference on
Fairness, Accountability, and Transparency,
Hasselbring, W. & Reussner, R. (2006). Toward
1475-1493.
trustworthy software systems. Computer,
https://doi.org/10.1145/3630106.3658984
39(4), 91–92.
https://doi.org/10.1109/MC.2006.142 Zhang, B., Liang, P., Zhou, X., Ahmad, A., &
Waseem, M. (2023). Demystifying Practices,
Jaworski, M., & Piotrkowski, D. (2023). Study of
Challenges and Expected Features of Using
software developers' experience using the
GitHub Copilot. International Journal of
GitHub Copilot Tool in the software
Software Engineering and Knowledge
development process. ArXiv,
Engineering, 33(1) 1653-1672.
abs/2301.04991.
https://doi.org/10.1142/S021819402341004
https://doi.org/10.48550/arXiv.2301.04991
8
Kedar, S. (2024). Tabnine vs. GitHub Copilot.
Tabnine. Retrieved on July 26, 2024, from
https://www.tabnine.com/blog/tabnine-
versus-GitHub-copilot/

©2024 ISCAP (Information Systems and Computing Academic Professionals) Page 8

https://iscap.us/proceedings/
2024 Proceedings of the ISCAP Conference ISSN: 2473-4901
Baltimore, MD v10 n6119

Appendix 1

©2024 ISCAP (Information Systems and Computing Academic Professionals) Page 9

https://iscap.us/proceedings/