Vol. 8, No.

3; December 2024
E-ISSN 2579-7484
Pages 685-700
DOI: http://dx.doi.org/10.29408/veles.v8i3.27700

Exploring High School EFL Teachers' Experiences with

Magic School AI in Lesson Planning: Benefits and Insights

*1Endang Setyaningsih, 1Muhammad Asrori, 1Ngadiso, 1Sumardi, 1Hasan Zainuri,

2Yuni Hariyanti
1Universitas Sebelas Maret, Surakarta, Indonesia
2Universitas Pembangunan Nasional Veteran Jakarta, Indonesia

*Correspondence:
[email protected]

Submission History:
Submitted: October 07, 2024
Revised: December 05, 2024
Accepted: December 10, 2024

This article is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Abstract
Advancements in artificial intelligence (AI) are reshaping education, yet its role in lesson planning warrants
further exploration. This study examines how Indonesian high school English as a Foreign Language (EFL)
teachers integrate MagicSchool AI, a generative AI tool, into lesson planning, revealing opportunities and
challenges. Data from 38 teachers, collected through post-workshops open-ended questionnaires, and focus
group interviews, highlight AI's dual potential: while it accelerates lesson planning, supports differentiated
instruction, and fosters creativity, it also necessitates critical teacher oversight, effective prompt-engineering
skills, and significant adaptation to align with classroom dynamics. Teachers emphasized AI’s value in reducing
workloads and generating diverse instructional ideas but noted that its outputs often lack readiness for
immediate use, requiring refinement to ensure pedagogical soundness. These findings underscore the
importance of targeted training and professional development to equip teachers with the skills to navigate AI
tools effectively. This study emphasizes the critical balance between human expertise and technological
efficiency by framing AI as a collaborative ally rather than a replacement for educators. It contributes to the
growing discourse on AI in education, offering insights into optimizing human-AI collaboration for innovative,
contextually relevant, and impactful teaching practices. This research invites educators and policymakers to
envision AI as a transformative force, capable of bridging innovation with pedagogical integrity to shape the
future of education.
Keywords: Artificial intelligence in Education (AIEd), EFL, generative AI, lesson planning

INTRODUCTION
The rapid advancement and widespread adoption of artificial intelligence (AI) across
various domains, including education, have sparked increasing interest among English
Language Teaching (ELT) researchers in uncovering its multifaceted impacts. Although
studies on AI in education remain limited, their numbers steadily rise. Crompton and Burke

685
Voices of English Language Education Society Vol. 8, No. 3; December 2024

(2023) observed a significant surge in AI-related publications between 2021 and 2022, with
numbers doubling or tripling compared to previous years, solidifying AI in education (AIEd)
as a prominent trend in recent educational research. These studies have explored diverse
applications of AI in learning and teaching, ranging from tutoring to automated assessments.
However, as González-Calatayud et al. (2021) noted, early studies up to 2020 largely
emphasized technical aspects of AI, often neglecting the underlying pedagogical frameworks.
More recent research has shifted focus, exploring not only the practical
implementation of AI but also addressing ethical considerations and integrity in its use
(Flora, 2024; Sharadgah & Sa’di, 2022). For instance, Flora (2024) raised concerns about AI’s
potential to compromise academic integrity. Marzuki et al. (2023) highlighted its positive
impact on students’ EFL writing content and organization, albeit with noted challenges. The
benefits of AI in education are widely acknowledged, particularly its ability to improve time
efficiency in lesson planning and facilitate personalized learning experiences for diverse
students (Jauhiainen & Guerra, 2023; Hasheem et al., 2024; van, 2023; Yilmaz, 2024). These
findings underscore AI's transformative potential while emphasizing the need for critical
and ethical integration into educational practices.
Lesson planning is a critical component of effective instruction, serving as a roadmap
for achieving learning objectives. A well-crafted lesson plan includes objectives, instructional
materials, activities, and assessments, often guided by national and local policies. Despite its
importance, many teachers struggle to produce high-quality lesson plans, as observed in our
workshops and teacher professional education programs (PPG). Common challenges include
incoherence, excessive assessment demands, and poor-quality language inputs. These
difficulties are not unique to Indonesia; Sharif et al. (2023) reported similar findings among
Malaysian EFL teachers, suggesting the need for further support in lesson planning. The
complexity of writing EFL lesson plans lies in addressing diverse student needs, balancing
culturally appropriate pedagogy, and managing pressures like high-stakes testing (Loh &
Liew, 2016).
Some previous studies have explored the use of AI in education, particularly in lesson
planning, revealing its potential and challenges. Sharif et al. (2023), in their analysis of
Malaysian EFL teachers, identified a moderate level of lesson planning competence and
emphasized the need for additional support. Writing effective EFL lesson plans remains
complex, requiring teachers to navigate culturally appropriate pedagogies, high-stakes
testing pressures, and students' diverse language proficiency, interests, and motivations
(Loh & Liew, 2016). Addressing these challenges, Octavio et al. (2024) found that ChatGPT
significantly supported Spanish EFL teachers by streamlining lesson design and learner
assessment, while Hasheem et al. (2024) highlighted its role in reducing planning time and
mitigating burnout through tailored lesson creation. Yeh (2024) further demonstrated AI’s
adaptability in creating interactive and differentiated learning materials, which catered to
diverse learner needs and enhanced teachers’ professional competencies. However, these
studies also underscore the critical need for prompt engineering and critical thinking skills
to ensure the pedagogical relevance of AI-generated outputs (Karpouzis, 2024; Kehoe,
2023). This body of research highlights the promise of AI as a tool to address lesson planning
complexities while reinforcing the importance of professional development to maximize its
potential.

686
Voices of English Language Education Society Vol. 8, No. 3; December 2024

While previous studies have highlighted the potential of AI in education, particularly

for lesson planning, several critical gaps remain. Much of the existing research, including
works by Sharif et al. (2023) and Yeh (2024), focuses on general benefits such as time
efficiency, adaptability, and workload reduction. However, these studies often neglect how
AI integrates with specific pedagogical frameworks, particularly in diverse and
underexplored teaching contexts. Furthermore, the emphasis on technical aspects, such as
prompt engineering, frequently overlooks broader pedagogical and contextual challenges
faced by teachers, especially in regions like Indonesia. While studies by Octavio et al. (2024)
and Hasheem et al. (2024) highlight AI’s utility in lesson planning, they provide limited
insights into how teachers critically evaluate and adapt AI-generated materials to address
learner diversity and dynamic classroom environments. Additionally, there is a lack of
research on the long-term implications of AI use on teachers’ professional development and
its impact on student outcomes.
This study seeks to address these gaps by focusing on the integration of MagicSchool
AI into lesson planning by Indonesian EFL teachers, a context that has received limited
attention in AI education research. Unlike previous studies that prioritize technical
functionalities, this research emphasizes the interplay between AI capabilities and teachers’
pedagogical expertise. By examining how teachers critically evaluate and adapt AI-generated
lesson materials to align with classroom needs and learner diversity, the study contributes
novel insights into the practical application of AI in lesson planning. Furthermore, it explores
the potential long-term effects of AI use on teachers’ professional competencies and its role
in enhancing student learning outcomes, bridging the gap between technical innovation and
pedagogical soundness.

METHOD
Research Design
This study employed an exploratory case study design to investigate the experiences
of 38 high school English as a Foreign Language (EFL) teachers in Central Java, Indonesia, as
they integrated MagicSchool AI into their lesson planning. This approach was chosen
because it allows for an in-depth examination of a relatively new and under-researched
phenomenon, capturing the teachers' unique challenges, strategies, and perceptions within
their specific educational and cultural context. Exploratory case studies are particularly
valuable for understanding complex phenomena in real-life contexts, especially when the
boundaries between the phenomenon and its context are unclear (Yin, 2018; Merriam &
Tisdell, 2015).
Participant Recruitment
Participants were recruited using convenience sampling, a method often employed in
exploratory studies for its practicality in accessing willing respondents (Etikan et al., 2016).
The target group comprised members of the district teacher forum (MGMP) with diverse
teaching backgrounds, ranging from 5 to more than 15 years of experience. All participants
held at least a bachelor's degree in English Education, with some having completed or
pursuing a master's degree. Table 1 summarizes the demographic and professional profiles

687
Voices of English Language Education Society Vol. 8, No. 3; December 2024

of the respondents, illustrating a representative diversity in gender, teaching experience, and

professional development engagement.
Table 1. Respondent's Identity
Gender Male Female
Educational Background
S1 5 28
S2 1 4
Years of teaching
5-7 years 1 3
7-10 years 3 21
10-13 years 2 4
13-15 years 0 1
More than 15 years 0 3
TPD Frequency
Often 0 4
Sometimes 4 20
Rare 2 8

Data Collection
Data were collected through an online open-ended questionnaire and focus group
interviews after a workshop on AI-assisted lesson planning. The questionnaire consisted of
two parts: the first collected demographic and professional background information, while
the second explored participants' experiences and reflections on using MagicSchool AI in
lesson planning. Focus group interviews, a widely used qualitative method, provided more
profound insights into participants' shared and individual experiences (Krueger & Casey,
2015). Three focal respondents were selected based on the following criteria: (1)
representation of different gender groups, (2) variation in teaching experience, and (3)
willingness and ability to articulate their experiences. The profiles of these focal respondents
are presented in Table 2.
Table 2. Focal Respondents
Lea Rosa Tony
English Education, BA English Education, BA English Education, BA
15 years of teaching 5 years of teaching More than 15 years of
experience experience teaching experience
Frequently join TPD Occasionally join TPD Occasionally join TPD

Stimulated Recall Interviews (SRI) were integrated into the focus group sessions to
enhance the depth of qualitative data. SRI has been recognized for its effectiveness in
capturing participants’ reflective thinking by revisiting specific events (Stough, 2001; Lyle,
2003). This method encouraged respondents to detail their cognitive and emotional
processes during AI integration, offering a rich dataset for analysis.

688
Voices of English Language Education Society Vol. 8, No. 3; December 2024

Data Analysis
Thematic analysis was employed to analyze the collected data systematically,
following established guidelines by Braun and Clarke (2006) and Miles et al. (2014). The
process began with data immersion, involving repeated readings of transcripts to gain a
comprehensive understanding of the content. Data condensation followed, removing
irrelevant or redundant statements and paraphrasing key ideas for clarity. The first coding
cycle used in-vivo coding, extracting keywords and phrases directly from participants'
responses to preserve the authenticity of their perspectives. These codes were tabulated and
clustered during the second coding cycle, where emerging patterns and themes were
identified. Iterative clustering and evaluations ensured the reliability and validity of the
identified themes.

FINDING AND DISCUSSION

The thematic analysis revealed four major themes representing the experiences of
EFL high school teachers using MagicSchool AI in lesson planning: (1) time efficiency, (2)
differentiation, (3) resourcefulness, and (4) challenges and lessons. These themes illustrate
generative AI's dual potential and limitations in supporting teaching practices.
Time Efficiency in Material and Assessment Preparation
Time efficiency emerged as a significant theme in the experiences of EFL teachers
using MagicSchool AI for lesson planning. The respondents highlighted how generative AI
streamlined the process of creating instructional materials and assessments, reducing the
time spent on traditionally labor-intensive tasks. For many teachers, the structured design
of MagicSchool AI provided a user-friendly interface that simplified the workflow, especially
compared to other AI tools like ChatGPT. However, the extent of these benefits varied based
on individual familiarity with AI tools and contextual factors such as internet connectivity.
The findings on time efficiency are summarized in Table 3, which categorizes respondents'
insights into material preparation, assessment design, and the challenges faced. These
categories offer a comprehensive overview of how MagicSchool AI-enhanced efficiency while
highlighting teachers' limitations in specific contexts.
Table 3. Key Findings on Time Efficiency
Sub-theme Respondent Insights Key Insights
Time efficiency - Lea: Found MagicSchool AI faster than MagicSchool AI significantly
in material ChatGPT due to structured prompts; reduces material preparation
preparation reduced holiday workload for material time for teachers familiar with its
preparation. structured fields.
- Tony: First-time user; found the
process intuitive and quick with pre-
designed fields.
Time efficiency - Malika: AI excels in generating diverse AI tools expedite assessment
in assessment question types quickly. creation, particularly for
design - Yaya: AI-generated assessments save generating varied and complex
time by providing varied difficulty tasks.
levels.

689
Voices of English Language Education Society Vol. 8, No. 3; December 2024

Challenges in - Rosa: Requires follow-up prompting to Challenges include poor internet

time efficiency refine outputs, reducing overall access and additional time
efficiency. required for refining AI-
- Saleh: Poor internet access makes AI generated outputs.
impractical for rural areas.
- Yasin: Limited connectivity increases
time and costs for AI use.

Time efficiency emerged as a central theme in the experiences of teachers using

MagicSchool AI for lesson planning, underscoring its potential to transform the traditionally
labor-intensive process of creating instructional materials and assessments. This finding
aligns with prior research emphasizing the role of generative AI in automating repetitive
tasks, thereby freeing up teachers’ time for more strategic and pedagogical work (Luckin et
al., 2016; Popenici & Kerr, 2017). The respondents in this study largely acknowledged the
tool's ability to streamline their workflows, particularly through structured prompts and
intuitive interfaces. For example, Lea, a focal respondent, observed that MagicSchool AI
reduced her holiday workload by providing a faster and more organized method of material
preparation compared to her previous reliance on ChatGPT. "It was faster for me to do
prompting because basic guides for prompting are already provided." This reflects findings by
Huang (2023), who noted that AI-generated outlines offer a foundation for educators to build
upon, emphasizing customization over starting from scratch.
The tool’s effectiveness in assessment design was another prominent aspect of time
efficiency. Respondents like Malika and Yaya highlighted how MagicSchool AI enabled them
to generate diverse question types with varying difficulty levels quickly. "AI excels in
generating diverse question types quickly," noted Malika, while Yaya added, "It saves time! I
just prepare a good prompt and text, and the AI creates questions that cater to different skill
levels." These observations corroborate findings by Holmes et al. (2019), who argue that AI-
driven tools enhance efficiency by automating complex processes, such as designing
differentiated assessments. Furthermore, Floris et al. (2024) emphasize that while AI tools
offer speed and variety, their outputs require alignment with specific pedagogical goals,
reinforcing the need for teacher oversight.
Despite these advantages, significant challenges emerged, tempering the perceived
efficiency of MagicSchool AI. Rosa pointed out that follow-up prompting was often necessary
to refine AI-generated content, occasionally reducing the expected time-saving benefits: "I
often have to ask the AI to simplify the words, add details, or adjust the length. In the end, it is
not as fast as I expected." This challenge reflects broader limitations of generative AI tools, as
highlighted by Tzirides et al. (2023), who caution that the quality of AI outputs depends
highly on the specificity and clarity of user inputs. Additionally, teachers in rural areas, such
as Saleh and Yasin, reported that poor internet connectivity hindered their ability to leverage
the tool effectively. Saleh lamented, "With limited internet access, working with AI takes too
much time. It is easier to do things the traditional way." At the same time, Yasin added, "I have
to go downtown for good internet access. But that means more time, more money, and more
energy." These findings resonate with the work of Sharadgah and Sa’di (2022), who

690
Voices of English Language Education Society Vol. 8, No. 3; December 2024

emphasize that infrastructural barriers can exacerbate the digital divide, limiting equitable
access to AI tools.
From a pedagogical perspective, the respondents’ experiences with MagicSchool AI
highlight the evolving role of educators in integrating AI into lesson planning. While tools
like MagicSchool AI streamline preparatory tasks, they also demand critical engagement
from teachers to adapt and refine the generated content for specific classroom contexts. This
duality supports the argument by Wang et al. (2024) that generative AI thrives when coupled
with substantial human involvement, ensuring that the outputs are both pedagogically
sound and contextually relevant. Furthermore, the requirement for follow-up prompting
underscores the importance of teacher training in prompt engineering, as the ability to craft
effective prompts directly impacts the quality of AI-generated materials (Kehoe, 2023; Floris
et al., 2024).
The implications of these findings are significant, underscoring several critical areas
for improvement and adaptation in the use of AI tools in education. First, they highlight the
necessity of targeted professional development programs that equip teachers with the skills
to optimize AI tools effectively. Such programs should address not only technical aspects,
such as prompt engineering, but also focus on critical evaluation skills to ensure the
pedagogical soundness of AI-generated outputs (Huang, 2023; Tzirides et al., 2023). Second,
addressing systemic barriers, such as reliable internet access, is crucial to ensuring equitable
access to these technological advancements, allowing all educators to benefit from the
potential of generative AI. Finally, the findings emphasize a transformative shift in lesson
planning practices, where generative AI tools act as collaborators, enhancing the educator's
role rather than replacing their expertise. This aligns with the insights of Luckin et al. (2016),
who argue that the integration of AI in education should support, rather than diminish, the
creative and critical dimensions of teaching.
While MagicSchool AI demonstrates substantial time-saving benefits in lesson
planning, its effectiveness hinges on multiple factors, including user proficiency,
infrastructural support, and the adaptability of AI-generated outputs to specific classroom
contexts. These findings contribute to the growing discourse on integrating AI in education,
reinforcing the importance of a balanced partnership between technology and pedagogy. By
recognizing both the opportunities and limitations of AI tools, educators can harness their
potential to streamline processes and enhance efficiency while preserving the essential
creative and critical elements of teaching.
Possible Differentiation
Beyond time efficiency in creating instructional materials and assessments,
differentiation emerged as another prominent theme in the respondents' experiences with
MagicSchool AI. Many teachers noted that the tool made it more feasible to level texts and
tailor content to address the diverse needs of their students. Rosa, for instance, highlighted
how MagicSchool AI-enabled her to create vocabulary-specific reading texts at varying levels
of complexity, a task she had previously found almost impossible due to workload
constraints. She explained:
"MagicSchool helps me create and vary reading texts based on the vocabulary that I
want students to learn. I often used authentic materials, but some of my students cannot
understand them. Also, sometimes the authentic texts do not contain the vocabulary that

691
Voices of English Language Education Society Vol. 8, No. 3; December 2024

I want to highlight. I like it when the tool not only uses the target vocabulary that I input
but also allows me to create different levels of complexity. Yes, it takes two steps, but it’s
certainly something I knew I should do but had never done because it was so difficult."
(Rosa, interview).
This insight aligns with Holmes et al. (2019), who argued that AI tools' ability to
personalize content to meet individual learner needs enhances motivation and learning
outcomes. Differentiation, as a pedagogical strategy, is particularly vital in classrooms with
diverse student abilities and preferences (Tomlinson, 2017). MagicSchool AI facilitates this
process by automating the creation of customized materials, allowing teachers to focus on
refining their teaching approaches to support varied learner profiles.
Lea further elaborated on how Gen-AI enabled her to diversify her teaching activities,
such as generating songs and jokes for use as icebreakers and engaging lesson introductions.
"I have learners with different learning styles and preferences, so varying ways of teaching and
delivery, in addition to leveling the instructional material, are considered beneficial for my
students," she stated. This practical application of Gen-AI reflects the growing emphasis on
inclusivity in education, where technology supports teachers in addressing multiple learning
modalities (Heacox, 2012). By leveraging AI, educators can cater to visual, auditory, and
kinesthetic learners more effectively, enhancing overall classroom engagement.
These findings align with the broader capabilities of AI to foster personalized learning
environments. Shandu et al. (2024) emphasized the adaptability of AI tools in tailoring
educational content to meet the needs of individual learners, reducing the one-size-fits-all
approach that often dominates traditional teaching. Moreover, Yeh (2024) highlighted that
generative AI empowers teachers to create interactive and adaptive materials, which not
only improve student engagement but also help teachers manage differentiated instruction
with greater ease.
Incorporating differentiation through AI also addresses a key challenge in education:
workload management. Teachers often struggle to implement differentiated strategies due
to time and resource limitations (VanTassel-Baska & Hubbard, 2016). By automating tasks
such as text leveling and content variation, MagicSchool AI reduces this burden, allowing
educators to focus on higher-order instructional strategies. Rosa noted, "It’s something I
knew I should do but had never done because it was so difficult. But using this AI, it’s a lot
easier."
Furthermore, differentiation through AI reflects a broader shift in teaching practices,
where technology bridges the gap between pedagogical challenges and innovation. The
ability to adapt content to individual learners not only supports inclusivity but also aligns
with the principles of Universal Design for Learning (UDL), which advocate for flexible and
personalized approaches to teaching (Nave, 2021). Gen-AI enables teachers to create
tailored content quickly, providing a more equitable and effective learning environment.
However, while Gen-AI’s role in differentiation is promising, its effectiveness depends
on teachers' ability to evaluate and adapt the generated outputs critically. As Floris et al.
(2024) highlighted, the pedagogical soundness of AI-generated materials requires human
oversight to ensure relevance and alignment with curriculum goals. This underscores the
importance of professional development programs that equip educators with the skills to
optimize AI tools effectively (Sharadgah & Sa’di, 2022).

692
Voices of English Language Education Society Vol. 8, No. 3; December 2024

MagicSchool AI has proven to be a critical enabler for differentiation, offering

educators a practical means to address diverse student needs while managing their
workload. Bridging technological innovation with pedagogical effectiveness supports
dynamic, student-centered teaching methodologies that enhance motivation and learning
outcomes. As generative AI continues to evolve, its role in fostering inclusivity and
personalization in education will likely become even more integral.
Resourcefulness
After attending the workshop, all focal respondents reported utilizing MagicSchool AI
multiple times to enhance their teaching practices. Rosa noted that while the tool provided
interesting ideas for class activities, she often needed to adjust them to fit her students'
specific needs. Similarly, Lea described how Gen-AI served as a valuable teaching resource,
prompting her to revisit and implement instructional strategies she had not used since the
early stages of her career. She reflected:
"I tend to be pragmatic in choosing the learning activities that are doable in class by
considering my students’ characteristics. AI provided some ideas that I think I can use. I
have been teaching for more than 15 years now, and I think I got accustomed to using
the same way to teach. I teach based on the book, asking students to do the exercises in
the book. When I asked AI to create activities, I got some that I think I want to try. They
are actually not new activities—I learned some of those back in my teacher training, but
I think I have forgotten them and never really used them except in my early career as a
teacher." (Lea, Interview)
Lea's experience underscores how AI tools can rejuvenate educators' instructional
approaches by reintroducing previously learned but underutilized teaching methods. This
phenomenon aligns with findings by Urmeneta and Romero (2024), who suggest that AI
stimulates expansive learning by encouraging educators to revisit and adapt established
pedagogical strategies. Additionally, AI-powered tools have been shown to enhance learning
environments by providing diverse instructional resources and facilitating personalized
learning experiences, supporting educators in delivering tailored and effective instruction
(Mishra et al., 2024).
Integrating AI in educational contexts also empowers teachers by expanding their
repertoire of instructional strategies. Trust and Maloy (2019) emphasize that digital tools
like MagicSchool AI catalyze teacher innovation, allowing educators to break away from
routine methods and explore more dynamic and engaging teaching approaches. Similarly,
Rosa’s observations reflect the flexibility of AI tools in adapting to specific classroom
contexts. While she acknowledged the necessity of modifying AI-generated ideas, she found
the suggestions invaluable as a foundation for designing tailored classroom activities. This
adaptability mirrors findings by Shandu et al. (2024), who highlighted AI's role in generating
diverse and contextually relevant resources, enabling teachers to address varied learning
needs.
However, adopting AI requires critical engagement to balance technological
efficiency with creative autonomy. Mishra et al. (2024) caution that while AI-generated
content can inspire innovative teaching practices, educators must critically evaluate and
refine these outputs to preserve authenticity and personalization. This sentiment echoes the

693
Voices of English Language Education Society Vol. 8, No. 3; December 2024

findings by Floris et al. (2024), who argue that AI tools can only complement teaching
effectively when paired with human oversight to ensure pedagogical soundness.
MagicSchool AI has proven to be a valuable resource for enhancing teacher creativity
and resourcefulness. By reintroducing forgotten strategies and generating new ideas, AI
tools support educators in creating engaging and tailored learning experiences. However,
their effectiveness hinges on educators' ability to critically adapt AI-generated content to
align with students' unique needs and learning contexts. This collaborative dynamic
between AI and teachers fosters a more innovative and adaptable teaching environment,
positioning technology as a supportive ally rather than a replacement for human expertise.
Challenges and lessons
While generative AI tools like MagicSchool have demonstrated their potential to
streamline lesson planning and inspire creativity, participants identified several challenges
that must be addressed to optimize their effectiveness. These challenges, including the
critical importance of precise prompting, the limitations of AI-generated outputs, and the
essential role of teacher oversight, highlight the nuanced nature of AI integration in
education. Respondents also emphasized that efficiency in lesson planning extends beyond
time-saving to include producing high-quality, contextually appropriate materials. Table 4
summarizes these challenges and lessons, providing key respondent insights alongside
actionable takeaways. This overview offers valuable guidance for educators and
stakeholders aiming to harness the potential of AI tools while addressing their inherent
limitations.
Table 4. Key Challenges and Lessons in Integrating AI Tools for Lesson Planning
Challenge/Lesso Key Respondent Insights Key Insights
n
Importance of - Tony: "Writing lesson plans will be Effective prompting ensures
Prompting effective if we have the right prompt. relevant and accurate AI outputs.
Mastering prompting is key when we
use AI."
Limitations of AI - Lea: Found some AI-generated texts AI-generated content often
Outputs unsuitable in vocabulary and requires refinement to meet
complexity; editing required specific classroom needs.
additional time.
- Rosa: Needed to simplify or modify AI-
generated outputs, adding to the
workload.
Role of Teacher - Rosa: "Teachers have to be flexible and Human oversight ensures
Oversight adaptive toward class dynamics, alignment with pedagogical and
something AI cannot cope with." contextual goals.
- Lea: AI limitations forced her to revisit
fundamental teaching principles.
Efficiency Beyond - Rosa: Redefined efficiency as the Efficiency extends beyond time-
Time-Saving ability to produce quality materials, saving to producing high-quality
not just save time. materials.

694
Voices of English Language Education Society Vol. 8, No. 3; December 2024

Lessons for - Kehoe (2023): Teachers must AI integration requires training

Effective supplement AI outputs with their in critical evaluation and
Integration insights. collaborative practices.
- Wang et al. (2024): Generative AI
thrives with human involvement and
judgment.

Integrating generative AI tools like MagicSchool AI into lesson planning has opened
new opportunities for efficiency and creativity. However, the findings reveal several
challenges educators must navigate to optimize their use. Participants emphasized the
importance of precise prompting, limitations in AI-generated outputs, and the critical role of
teacher oversight, offering valuable insights into the dynamics of human-AI collaboration in
education.
Prompting emerged as a fundamental skill for achieving effective results with AI tools.
Tony highlighted that mastering prompting is essential, noting that "writing lesson plans will
be effective if and only if we have the right prompt. Mastering prompting is key when we use
AI." This sentiment aligns with the conclusions of Tzirides et al. (2023), Floris et al. (2024),
and Mishra et al. (2024), who collectively emphasize that the quality of AI outputs is directly
proportional to the quality of user inputs. Without precise prompts, AI-generated materials
risk being vague, irrelevant, or impractical for classroom use. These findings reinforce the
need for professional development programs that enhance educators’ prompt engineering
skills to ensure AI-generated materials meet pedagogical standards.
The findings also revealed limitations in the outputs generated by AI tools. Lea and
Rosa expressed concerns about the suitability of vocabulary and contextualization in AI-
generated texts. Lea explained, "I still have to edit and sometimes replace all the generated
text, and that takes a lot of time." Similarly, Rosa noted the need to simplify or modify outputs
to align with her students’ needs. These challenges are consistent with insights from Kehoe
(2023), Wang et al. (2024), and Holmes et al. (2019), who argue that while AI tools can
expedite material creation, they often require significant refinement to address classroom
diversity and contextual appropriateness. Such findings highlight the complementary role of
AI and educators, where human oversight ensures that AI-generated content aligns with
learning objectives and classroom realities.
Teacher oversight was another significant theme in the findings. Rosa emphasized
that "teachers have to be flexible and adaptive toward class dynamics, something AI cannot
cope with." This observation echoes the conclusions of Floris et al. (2024), Luckin et al.
(2016), and Tzirides et al. (2023), who collectively underscore the indispensable role of
educators in preserving the pedagogical and contextual integrity of AI-generated materials.
By critically evaluating and adapting AI outputs, teachers ensure that the generated content
meets the specific needs of their students and aligns with curriculum goals. Lea’s reflection
on AI limitations highlighted how these tools can prompt educators to revisit fundamental
teaching principles, demonstrating the iterative and complementary relationship between
AI and human expertise.
While efficiency was a key advantage of AI tools, the findings suggest that it extends
beyond mere time-saving. Rosa redefined efficiency as the ability to produce quality
materials within a reasonable timeframe, stating that "the time efficiency definition is no

695
Voices of English Language Education Society Vol. 8, No. 3; December 2024

longer limited to the quantitative amount of time to produce but the amount of time used to
produce quality products." This perspective aligns with the work of Luckin et al. (2016), Wang
et al. (2024), and Urmeneta and Romero (2024), who emphasize that actual efficiency in
education involves creating impactful and contextually relevant teaching materials. AI tools
like MagicSchool AI reduce the time required for initial material creation but demand
additional effort from educators to refine and adapt these materials for optimal use.
The lessons from these findings underscore the importance of professional
development programs that prepare teachers to use AI tools effectively. Kehoe (2023), Wang
et al. (2024), and Sharadgah and Sa’di (2022) collectively highlight the need for teachers to
supplement AI-generated content with their expertise to make informed pedagogical
decisions. Professional development should focus on enhancing prompt engineering skills,
fostering critical evaluation capabilities, and addressing ethical considerations in AI use.
Such programs would ensure that teachers are not only proficient in using AI tools but are
also equipped to integrate them into their pedagogical practices in a way that enhances
educational outcomes.
The interplay between AI tools and teacher expertise raises critical questions about
the balance between automation and human input in education. Over-reliance on AI risks
diminishing creativity and critical thinking in lesson planning, as noted by participants and
supported by Tzirides et al. (2023) and Floris et al. (2024). Future research should explore
the optimal balance of human-AI collaboration, examining how educators can leverage AI to
enhance teaching effectiveness while preserving essential human elements. Addressing
systemic challenges, such as infrastructural limitations, is also crucial to ensure equitable
access to AI tools, enabling educators in diverse contexts to benefit from technological
advancements.
These findings highlight the transformative potential of AI in lesson planning while
emphasizing the indispensable role of educators. By addressing limitations, refining
integration strategies, and investing in teacher training, AI tools like MagicSchool AI can
become powerful allies in education. However, their success relies on a balanced partnership
between human expertise and technological innovation, ensuring that lesson planning
remains efficient and pedagogically effective. This balance will ultimately create meaningful
and impactful learning experiences guided by the collaborative efforts of educators and AI
tools.

CONCLUSION
This study highlights the transformative potential of generative AI tools like
MagicSchool in enhancing lesson planning and addressing the critical challenges they
present. The findings reveal that effective AI integration requires precise prompting,
significant teacher oversight, and adaptability to contextual classroom needs. While AI
tools demonstrate clear advantages in streamlining material creation and offering diverse
instructional ideas, their outputs often require refinement to meet pedagogical standards,
underscoring the indispensable role of educators in maintaining quality and relevance.
Notably, the concept of efficiency was reframed by respondents to emphasize the
production of high-quality, contextually appropriate materials over mere time-saving.
These insights underscore the need for professional development that equips teachers with

696
Voices of English Language Education Society Vol. 8, No. 3; December 2024

technical skills in prompt engineering and critical evaluation, fostering a balanced and
effective human-AI partnership. By addressing these challenges, AI tools can be harnessed
not as replacements for educators but as collaborative allies, enabling innovative, efficient,
and impactful teaching practices. Future research should further explore optimal models
for human-AI collaboration and strategies to ensure equitable access, particularly in
resource-constrained educational contexts.

ACKNOWLEDGMENTS
We sincerely express our gratitude to the High-School English Teacher Forum in
Wonogiri for their enthusiastic participation and invaluable contributions to our workshop
and research. We also extend our heartfelt thanks to the Institute for Research and
Community Service (LPPM) at Universitas Sebelas Maret (UNS) for their unwavering support
and facilitation, which significantly contributed to the successful implementation of this
study and community service initiative.

REFERENCES
Ball, S., Kenny, A., & Gardiner, D. (1990). Literacy, politics, and the teaching of English. In I.
Goodson & P. Medway (Eds.), Bringing English to order (pp. 47–86). The Falmer Press.
Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative research in
psychology, 3(2), 77-101.
Crompton, H., & Burke, D. (2023). Artificial intelligence in higher education: The state of the
field. International Journal of Education Technology in Higher Education, 20(22).
https://doi.org/10.1186/s41239-023-00392-8
Ethics of Artificial Intelligence. (2024, September 26). UNESCO.
https://www.unesco.org/en/artificial-intelligence/recommendation-ethics
Etikan, I., Musa, S. A., & Alkassim, R. S. (2016). Comparison of convenience sampling and
purposive sampling. American journal of theoretical and applied statistics, 5(1), 1-4.
https://doi.org/10.11648/j.ajtas.20160501.11
Flora, D. B. (2024). Introducing AI in Precision Oncology. AI in Precision Oncology, 1(1), 1-2.
Floris, F. D. (2023). Unveiling teachers' perceptions: A journey into the use of artificial
intelligence tools in English language teaching. [Unpublished doctoral dissertation]. State
University of Malang.
Floris, F. D. (2024). AI ethics 101: Pedagogical. Modern English Teacher, 33(2).
Floris, F. D., Widiati, U., Renandya, W. A., & Busthomi, Y. (2024). Artificial intelligence in
English language teaching: Fostering joint enterprise in online communities. Journal of
English Educators Society, 9(1), 12–21.
Ghimire, P. R., Neupane, B. P., & Dahal, N. (2024). Generative AI and AI tools in English
language teaching and learning: An exploratory research. English Language Teaching
Perspectives, 9(2), 30–40.
González-Calatayud, V., Prendes-Espinosa, P., & Roig-Vila, R. (2021). Artificial intelligence for
student assessment: A systematic review. Applied Sciences, 11(12), 54–67.
https://doi.org/10.3390/app11125467
Hashem, R., Ali, N., Zein, F., Fidalgo, P., & Abu Khurma, O. (2023). AI to the rescue: Exploring
the potential of ChatGPT as a teacher ally for workload relief and burnout prevention.

697
Voices of English Language Education Society Vol. 8, No. 3; December 2024

Research and Practice in Technology Enhanced Learning, 19.

https://doi.org/10.58459/rptel.2024.19023
Heacox, D. (2012). Differentiating instruction in the regular classroom: How to reach and
teach all learners (Updated anniversary edition). Free Spirit Publishing.
Holmes, W., Bialik, M., & Fadel, C. (2019). Artificial intelligence in education: Promises and
implications for teaching and learning. Center for Curriculum Redesign.
Huang, W. (2023, July). The future of your lesson planning: When a culture lesson meets
artificial intelligence. The FLTMag.
Iskender, A. (2023). Holy or unholy? Interview with open AI’s ChatGPT. European Journal of
Tourism Research, 34. https://doi.org/10.54055/ejtr.v34i.3169
Jauhiainen, J. S., & Guerra, A. G. (2023). Generative AI and ChatGPT in school children’s
education: Evidence from a school lesson. Sustainability, 15(18), 14025.
https://doi.org/10.3390/su151814025
Johinke, R., Cummings, R., & DiLauro, F. (2023). Reclaiming the technology of higher
education for teaching digital writing in a post-pandemic world. Journal of University
Teaching & Learning Practice, 20(2), 01. https://doi.org/10.53761/1.20.02.01
Karpouzis, K. (2024). Plato’s Shadows in the Digital Cave: Controlling Cultural Bias in
Generative AI. Electronics, 13(8), 1457. https://doi.org/10.3390/electronics13081457
Kehoe, F. (2023). Leveraging generative AI tools for enhanced lesson planning in initial
teacher education at post primary. Irish Journal of Technology Enhanced Learning. Doi:
10.22554/ijtel.v7i2.124
Krueger, R. A., & Casey, M. A. (2015). Focus group interviewing. Handbook of practical
program evaluation, 506-534.
Loh, C. E., & Liew, W. M. (2016). Voices from the ground: The emotional labour of English
teachers’ work. Teaching and Teacher Education, 55, 267–278.
https://doi.org/10.1016/j.tate.2016.01.016
Luckin, R., Holmes, W., Griffiths, M., & Forcier, L. B. (2016). Intelligence unleashed: An
argument for AI in education. Pearson Education.
Lyle, W. A., & Jin, G. J. (2003). Laser in situ keratomileusis for consecutive hyperopia after
myopic LASIK and radial keratotomy. Journal of Cataract & Refractive Surgery, 29(5),
879-888. https://doi.org/10.1016/S0886-3350(03)00051-8
Marzuki, , Widiati, U., Rusdin, D., Darwin, , & Indrawati, I. (2023). The impact of AI writing
tools on the content and organization of students’ writing: EFL teachers’ perspective.
Cogent Education, 10(2). https://doi.org/10.1080/2331186X.2023.2236469
Merriam, S. B., & Tisdell, E. J. (2015). Qualitative research: A guide to design and
implementation. John Wiley & Sons.
Miles, M. B., Huberman, A. M., & Saldana, J. (2018). Qualitative data analysis: A methods
sourcebook. SAGE Publications.
Mishra, S., Rao, A., Krishnan, R., Ayyub, B., Aria, A., & Zio, E. (2024). Reliability, Resilience and
Human Factors Engineering for Trustworthy AI Systems. arXiv preprint
arXiv:2411.08981. https://doi.org/10.48550/arXiv.2411.08981
Nave, L. (2021). Universal design for learning: UDL in online environments: The WHAT of
learning. Journal of developmental education, 44(2), 30-32.
http://www.jstor.org/stable/45381107

698
Voices of English Language Education Society Vol. 8, No. 3; December 2024

Octavio, M. M., Argüello, M. V. G., & Pujolà, J. (2024). ChatGPT as an AI L2 teaching support: A
case study of an EFL teacher. Technology in Language Teaching & Learning, 6(1), 1–25.
https://doi.org/10.29140/tltl.v6n1.1142
Popenici, S. A., & Kerr, S. (2017). Exploring the impact of artificial intelligence on teaching
and learning in higher education. Research and Practice in Technology Enhanced
Learning, 12(1).
Sandhu, R., Channi, H. K., Ghai, D., Cheema, G. S., & Kaur, M. (2024). An introduction to
generative AI tools for education 2030. In Advances in Educational Technologies and
Instructional Design Book Series (pp. 1–28). https://doi.org/10.4018/979-8-3693-2440-
0.ch001
Sharadgah, T.A., Sa'di, R.A. (2022). A systematic review of research on the use of artificial
intelligence in English language teaching and learning (2015-2021): What are the
current effects? Journal of Information Technology Education (21). 337-377.
https://doi.org/10.28945/4999
Sharif, N. A.M., Damio, S. M., & Mashwani, H. U. (2023). Enhancing language acquisition: A
case study of TESL lesson plans in an international school. Malaysian Journal of Social
Sciences and Humanities (MJSSH). 8(10). https://doi.org/10.47405/mjssh.v8i10.2506
Stough, L. M. (2001). Using Stimulated Recall in Classroom Observation and Professional
Development. http://files.eric.ed.gov/fulltext/ED457214.pdf
Tomlinson, C. A. (2017). How to differentiate instruction in academically diverse classrooms.
Ascd.
Trust, T., & Maloy, R. W. (2019). Preparing teachers for the digital age: A framework for
integrating digital tools into teacher education. Educational Technology, 59(2).
Tzirides, A.O, Zapata G., Saini A., Searsmith D., Cope, B., Kalantzis, M., Castro V., Kourkoulou,
T., Jones J., da Silva, R.A., Whiting, J., and Kastania, N.P. (2023). Generative AI:
Implications and applications for education. Doi:
https://doi.org/10.48550/arXiv.2305.07605
Urmeneta, A., & Romero, M. (2024). Creative Applications of Artificial Intelligence in
Education (p. 181). Springer Nature.
Van den Berg G., du Plessis, E. (2023). ChatGPT and generative AI: Possibilities for Its
contribution to lesson planning, critical thinking and openness in teacher education.
Education Sciences. 13(10):998. https://doi.org/10.3390/educsci13100998
Wang, S., Wang, F., Zhu, Z., Wang, J., Tran, T., & Du, Z. (2024). Artificial intelligence in
education: A systematic literature review. Expert Systems with Applications.
https://doi.org/10.1016/j.eswa.2024.124167
Williyan, A., Fitriati, S. W., Pratama, H., & Sakhiyya, Z. (2024). AI as co-creator: Exploring
Indonesian EFL teachers’ collaboration with AI in content development. Teaching
English with Technology. 24(2), 5–21.
https://doi.org/10.56297/vaca6841/LRDX3699/RZOH5366
Yeh, H. C. (2024). The synergy of generative AI and inquiry-based learning: transforming the
landscape of English teaching and learning. Interactive Learning Environments, 1-15.
https://doi.org/10.1080/10494820.2024.2335491

699
Voices of English Language Education Society Vol. 8, No. 3; December 2024

Yılmaz, Ö. (2024). Personalised learning and artificial intelligence in science education:

Current state and future perspectives. Educational Technology Quarterly.
https://doi.org/10.55056/etq.744.
Yin, R. K. (2018). Case Study Research and Applications: Design and Methods (6th ed.).
Thousand Oaks, CA: Sage.

700