الحدود | استكشاف ممارسات التعلم الموجه ذاتياً من خلال ChatGPT بين متعلمي اللغة العربية غير الناطقين بها في إحدى الجامعات السعودية

خلاصة

[–>

Introduction:

[–>[–>

This study examined how non-native learners of Arabic at a Saudi university used ChatGPT as part of their self-directed learning practices. It focused on the role of ChatGPT in supporting Arabic language development and fostering motivation for independent learning.

[–>[–>

Methods:

[–>[–>

Adopting a qualitative research design, the study collected data through semi-structured, in-depth interviews with six learners from diverse ethnic backgrounds. The data were analyzed using thematic analysis to identify recurring themes and patterns in participants' experiences.

[–>[–>

Results:

[–>[–>

The findings revealed that ChatGPT played a meaningful role in supporting learners' Arabic language acquisition. Participants reported improvements in vocabulary, grammatical knowledge, and reading comprehension, highlighting the tool's value as a readily accessible learning resource. In addition, ChatGPT appeared to strengthen learners' motivation for autonomous learning by encouraging greater independence, promoting active engagement with the language, and allowing learners to tailor their learning experiences to their individual needs.

[–>[–>

Discussion:

[–>[–>

Overall, the study demonstrates the potential of AI-powered tools such as ChatGPT to enhance Arabic language learning and support the development of learner autonomy.

[–>

1 Introduction

Arabic is one of the most widely spoken languages in the world, with more than 400 million speakers across the Middle East, North Africa, and numerous diaspora communities. In addition to its status as an official language of international organizations such as the United Nations and UNESCO, Arabic occupies a unique place in global culture because it is the language of the Holy Qur'an and a central vehicle of Islamic scholarship and civilization. Over the past decade, interest in learning Arabic as a foreign language (AFL) has grown considerably. Students pursue Arabic for a variety of reasons, including academic study, career opportunities, religious purposes, and intercultural communication ([–>Alshehri et al., 2025;[–>[–>Ritonga et al., 2024;[–>[–>Soliman and Khalil, 2024).

على الرغم من هذا الاهتمام المتزايد، فإن تعلم اللغة العربية يمثل عددًا من التحديات لغير الناطقين بها. تتميز اللغة بنظام صرفي غني، وهياكل نحوية معقدة، وطبيعة ثنائية اللغة تتطلب من المتعلمين التنقل بين اللغة العربية الفصحى الحديثة والأصناف المنطوقة. بالإضافة إلى ذلك، فإن النص العربي ونظام الكتابة من اليمين إلى اليسار والمفردات الواسعة غالبًا ما تخلق صعوبات للمتعلمين، خاصة في القراءة والكتابة. ونتيجة لذلك، يحتاج العديد من الطلاب إلى دعم يمتد إلى ما هو أبعد من التدريس في الفصول الدراسية من أجل تطوير الكفاءة والحفاظ على التقدم المطرد ([–>[–>AlAwfi, 2024;[–>[–>Almelhes, 2024).

واستجابة لهذه التحديات، وجه معلمو اللغة والباحثون اهتمامهم بشكل متزايد إلى التعلم المعزز بالتكنولوجيا. لقد تم استخدام الأدوات الرقمية منذ فترة طويلة لتكملة تعليم اللغة، ولكن التطورات الأخيرة في الذكاء الاصطناعي (AI) قدمت إمكانيات جديدة لتجارب التعلم الشخصية والتفاعلية. على عكس منصات تعلم اللغة التقليدية التي تعتمد على التمارين الثابتة والاستجابات المبرمجة مسبقًا، يمكن للأنظمة القائمة على الذكاء الاصطناعي التكيف بسهولة أكبر مع احتياجات المتعلم الفردية وتقديم تعليقات فورية، مما يخلق فرصًا للممارسة والمشاركة المستمرة ([–>[–>Qpilat and Salam, 2023;[–>[–>Li et al., 2022).

ومن بين هذه التطورات، برز ChatGPT كأحد أدوات الذكاء الاصطناعي الأكثر مناقشة على نطاق واسع في التعليم. يعتمد ChatGPT على تقنية نماذج اللغة الكبيرة، ويمكنه إنشاء استجابات شبيهة بالردود البشرية، وإشراك المستخدمين في محادثات موسعة، وتقديم تفسيرات حول مجموعة واسعة من المواضيع. وقد اجتذبت هذه القدرات اهتمامًا كبيرًا من المعلمين والباحثين والطلاب على حدٍ سواء. في سياقات تعلم اللغة، يمكن لـ ChatGPT أن يخدم أدوارًا متعددة في وقت واحد: مدرس، وشريك محادثة، ومساعد كتابة، ومصدر للملاحظات. يمكن للمتعلمين طرح أسئلة حول القواعد واستكشاف معاني المفردات وطلب أمثلة لاستخدام اللغة وتلقي الدعم في مهام القراءة والكتابة في الوقت الفعلي ([–>[–>Kohnke et al., 2023;[–>[–>Jiang et al., 2022).

بالنسبة لمتعلمي اللغات الأجنبية، فإن إحدى الميزات الأكثر جاذبية في ChatGPT هي إمكانية الوصول إليه. يمكن للطلاب التفاعل مع النظام عندما يحتاجون إلى المساعدة، بغض النظر عن الوقت أو الموقع. تسمح هذه المرونة للمتعلمين بممارسة المهارات اللغوية خارج الفصل الدراسي وبالسرعة التي تناسبهم. تعتبر هذه الفرص ذات قيمة خاصة للأفراد الذين قد يكون لديهم وصول محدود إلى المتحدثين الأصليين أو التعليم الشخصي. ونتيجة لذلك، بدأ الباحثون في دراسة كيفية تأثير ChatGPT على نتائج تعلم اللغة، ومشاركة المتعلم، وسلوكيات التعلم المستقلة عبر البيئات التعليمية المختلفة ([–>[–>Boudouaia et al., 2024;[–>[–>ZainEldin et al., 2024).

2 Theoretical framework

This study is guided by the concepts of self-directed learning and learner autonomy, two perspectives that emphasize the active role learners play in managing their own educational development.

[–>[–>Knowles (1975) يصف التعلم الموجه ذاتيًا بأنه عملية يتحمل فيها المتعلمون مسؤولية تحديد احتياجاتهم التعليمية، وتحديد الأهداف، وتحديد الموارد، واختيار استراتيجيات التعلم، وتقييم النتائج. وبدلاً من الاعتماد كليًا على توجيهات المعلم، يصبح المتعلمون مشاركين نشطين في تشكيل تجاربهم التعليمية. وقد أصبح هذا المفهوم ذا أهمية متزايدة في السياقات التعليمية المعاصرة، حيث توفر التقنيات الرقمية وصولاً غير مسبوق إلى المعلومات وموارد التعلم.

ترتبط استقلالية المتعلم ارتباطًا وثيقًا بالتعلم الموجه ذاتيًا.[–>[–>Holec (1981) يُعرّف استقلالية المتعلم بأنها القدرة على تحمل مسؤولية قرارات التعلم الخاصة به. يحدد المتعلمون المستقلون ما يريدون تعلمه، وكيف ينوون تعلمه، وكيف سيقيمون تقدمهم.[–>[–>Benson (2013) يجادل بأن الاستقلالية مهمة بشكل خاص في تعلم اللغة الثانية والأجنبية لأن تطوير اللغة ذات المعنى غالبًا ما يتطلب الممارسة والتعرض والتفاعل بما يتجاوز التدريس الرسمي في الفصول الدراسية.

توفر وجهات النظر هذه أساسًا مفيدًا لفهم الدور المحتمل لـ ChatGPT في تعلم اللغة. من خلال الوصول الفوري إلى التوضيحات والتعليقات الشخصية وفرص الممارسة المستمرة، قد يساعد ChatGPT المتعلمين على التحكم بشكل أكبر في رحلات تعلم اللغة الخاصة بهم. وبدلاً من انتظار مساعدة المعلم، يمكن للمتعلمين طلب التوضيح بشكل مستقل، وتجربة استخدام اللغة، ومراقبة تقدمهم. وبهذا المعنى، يمكن النظر إلى ChatGPT ليس فقط كأداة تكنولوجية ولكن كميسر محتمل للتعلم الموجه ذاتيًا والمستقل.

[–>

2.1 ChatGPT and language learning

[–>[–>

The increasing interest in artificial intelligence has led researchers to reconsider how technology can support language learning both inside and outside the classroom. Among the various AI-driven innovations, conversational systems such as ChatGPT have attracted particular attention because they offer learners opportunities for interaction that extend beyond the static exercises typically found in conventional language-learning applications. Rather than simply delivering pre-programmed responses, ChatGPT can engage users in extended dialogue, respond to individual questions, and generate personalized feedback. These capabilities have prompted many educators and researchers to view the technology as a potentially valuable resource for language learning. At the same time, growing enthusiasm for AI has sometimes outpaced the empirical evidence, making it important to examine not only what ChatGPT can do, but also how learners actually use it and what educational value those interactions provide.

[–>[–>

Much of the existing research has highlighted the versatility of ChatGPT in supporting different aspects of language development. Learners commonly use the tool to obtain grammatical explanations, clarify unfamiliar vocabulary, practice conversational exchanges, and receive assistance with reading and writing tasks ([–>Kohnke et al., 2023;[–>[–>Jiang et al., 2022). تكمن جاذبيتها إلى حد كبير في سرعة استجاباتها ومرونة تجربة التعلم التي تقدمها. بدلاً من انتظار تعليقات المعلم أو البحث في مصادر تعليمية متعددة، يمكن للمتعلمين التفاعل مع ChatGPT كلما ظهرت أسئلة. وقد دفعت إمكانية الوصول هذه بعض العلماء إلى اقتراح أن الأنظمة التي تعمل بالذكاء الاصطناعي قد تساعد في سد الفجوات بين التدريس في الفصول الدراسية والدراسة المستقلة.

[–>

Writing has emerged as one of the most frequently studied areas of ChatGPT use. Foreign language learners often struggle with generating ideas, organizing content, and maintaining grammatical accuracy, particularly when writing independently. Research suggests that ChatGPT can support learners during various stages of the writing process by offering suggestions, identifying language errors, and providing feedback on clarity and organization ([–>Mohideen, 2024;[–>[–>Boudouaia et al., 2024). ومع ذلك، فإن القيمة المحتملة لهذا الدعم قد تكمن بشكل أقل في التصحيحات نفسها بقدر ما تكمن في كيفية تفاعل المتعلمين معها. إن مجرد قبول المراجعات التي ينشئها الذكاء الاصطناعي قد لا يفعل الكثير لتعزيز تطوير اللغة على المدى الطويل. من المرجح أن تظهر الفوائد التعليمية عندما يقوم المتعلمون بتقييم الملاحظات بشكل نقدي، ومقارنة البدائل، والتفكير في الخيارات اللغوية المقترحة. وبهذا المعنى، قد يعمل ChatGPT بشكل أكثر فعالية كسقالة للتعلم وليس كحل آلي لصعوبات الكتابة.

[–>

A similar pattern can be observed in vocabulary learning. Researchers have reported that learners use ChatGPT to obtain definitions, explore word usage, and generate contextualized examples. Such interactions can expose learners to language in ways that traditional vocabulary lists often cannot. Nevertheless, vocabulary acquisition is a gradual process that requires repeated exposure and meaningful use across different contexts. While ChatGPT can facilitate this process by generating varied examples and explanations, exposure alone does not guarantee learning. As several scholars have noted, the effectiveness of AI-mediated vocabulary practice depends largely on the extent to which learners actively engage with new lexical items rather than merely consuming information provided by the system.

[–>[–>

Beyond writing and vocabulary development, ChatGPT has been used to support reading comprehension and conversational practice. Learners frequently ask the system to summarize texts, explain difficult passages, or engage in simulated dialogues in the target language. These activities can create valuable practice opportunities, particularly for learners who have limited access to native speakers or individualized instruction. Yet it is important to recognize that interactions with AI differ fundamentally from authentic human communication. Although ChatGPT can simulate conversation remarkably well, it does not participate in communication as a social actor with genuine intentions, emotions, or cultural experiences. Consequently, while AI-based interactions may complement language practice, they cannot fully replicate the social and cultural dimensions of communication that are central to language use.

[–>[–>

The literature has also drawn attention to the motivational benefits associated with ChatGPT. Many learners describe the tool as approachable, readily available, and non-judgmental, characteristics that may lower anxiety and encourage experimentation with the target language ([–>ZainEldin et al., 2024;[–>[–>El Zahraa, 2025). وهذا مهم بشكل خاص في سياقات اللغات الأجنبية حيث قد يتردد المتعلمون في المشاركة بسبب المخاوف من ارتكاب الأخطاء. ومع ذلك، لا ينبغي النظر إلى الدافع على أنه نتيجة تلقائية لاستخدام التكنولوجيا. يمكن أن يتضاءل الحماس الأولي للتقنيات الجديدة بمرور الوقت، وقد يحكم المتعلمون في النهاية على قيمة الأداة وفقًا لمدى دعمها لأهدافهم التعليمية. ولذلك، من المرجح أن تعتمد المشاركة المستدامة ليس فقط على حداثة ChatGPT ولكن أيضًا على تصورات المتعلمين لفائدته وصلته باحتياجاتهم في تعلم اللغة.

[–>

From a theoretical perspective, ChatGPT appears to align closely with principles of self-directed learning and learner autonomy. Its availability enables learners to seek information independently, address gaps in their knowledge, and engage in self-paced practice. These features have led some researchers to argue that AI technologies can empower learners to take greater responsibility for their own learning. Yet autonomy involves more than having access to resources. As [–>Benson (2013) يقول أن التعلم المستقل يتطلب من المتعلمين اتخاذ قرارات مستنيرة، وتقييم المعلومات بشكل نقدي، وتنظيم عمليات التعلم الخاصة بهم. قد يؤدي الوصول إلى ChatGPT إلى خلق فرص للتعلم المستقل، ولكن تحقيق هذه الفرص يعتمد إلى حد كبير على قدرة المتعلمين على استخدام التكنولوجيا بشكل استراتيجي ونقدي.

[–>

At the same time, the limitations of ChatGPT have generated substantial discussion within the literature. One concern involves the possibility of inaccurate or fabricated information, often referred to as AI hallucinations. Although the system typically produces fluent and convincing responses, fluency should not be equated with accuracy. Learners who lack sufficient linguistic or subject knowledge may struggle to identify errors, potentially resulting in misunderstandings or the reinforcement of incorrect information ([–>Kasneci et al., 2023). يسلط هذا التحدي الضوء على أهمية تطوير مهارات القراءة والكتابة الرقمية الهامة إلى جانب استخدام تقنيات الذكاء الاصطناعي.

[–>

Another concern relates to learner dependence on AI-generated support. While immediate assistance can facilitate learning, excessive reliance on automated feedback may reduce learners' engagement with the cognitive processes that underpin language development. Language learning requires sustained effort, problem-solving, and reflection. If learners routinely delegate these processes to AI systems, they may become less active participants in their own learning. Consequently, the educational value of ChatGPT should not be measured solely by the quality of the responses it provides, but also by the extent to which it encourages meaningful learner engagement.

[–>[–>

Overall, the existing literature presents a more nuanced view of ChatGPT than either enthusiastic advocacy or outright skepticism suggests. The technology offers considerable potential to support language learning by expanding access to practice, feedback, and learning resources. However, its effectiveness cannot be understood independently of the learners who use it and the educational contexts in which it is employed. Rather than viewing ChatGPT as a substitute for teachers or established pedagogical practices, it may be more productive to regard it as a resource whose value depends on thoughtful, critical, and purposeful use. Such a perspective is particularly relevant for learners of Arabic as a foreign language, where questions concerning independent learning, language practice, and learner autonomy remain underexplored.

[–>[–>

2.2 Research gap

[–>[–>

The growing interest in artificial intelligence has generated an expanding body of research on its applications in language education. However, despite the increasing number of studies examining AI-assisted language learning, several important questions remain insufficiently explored. One notable issue concerns the linguistic contexts represented in the existing literature. Much of the current research has focused on English and other widely taught foreign languages, while Arabic has received comparatively limited attention. This imbalance is significant because Arabic presents learners with a set of linguistic challenges that differ in important ways from those associated with many other languages. Its complex morphology, diglossic nature, and distinctive writing system shape the learning experience in ways that may influence how learners interact with AI-based tools. As a result, findings from studies conducted in other language contexts cannot simply be assumed to apply to learners of Arabic.

[–>[–>

Another limitation concerns the settings in which AI-supported learning has been investigated. Existing studies have largely examined learner achievement, learner attitudes, or the integration of AI technologies within formal instructional environments. While these areas have contributed valuable insights, they provide only a partial picture of how learners engage with AI in their everyday learning practices. Language development often extends beyond the classroom, particularly in foreign-language contexts where learners must create opportunities for additional exposure and practice. Yet relatively little is known about how learners use tools such as ChatGPT independently, what purposes these tools serve in their learning routines, and how they contribute to ongoing language development outside teacher-directed environments.

[–>[–>

The experiences of learners studying Arabic in Saudi Arabia represent another area that remains underexplored. Saudi universities have long played a central role in providing Arabic-language education to international students and have increasingly embraced digital and emerging educational technologies. Despite this context, relatively few studies have examined how non-native learners of Arabic in Saudi higher education engage with AI-powered learning tools. Consequently, there is limited empirical evidence regarding how these learners incorporate ChatGPT into their learning practices, what types of language-learning activities they undertake through the platform, and how they evaluate its contribution to their linguistic development.

[–>[–>

A further issue is that much of the literature discusses artificial intelligence in broad terms, often treating AI as a single educational phenomenon. Such an approach can obscure important differences between specific technologies and how they are used by learners. ChatGPT, for example, differs from many previous language-learning tools because of its conversational capabilities, accessibility, and capacity to generate individualized responses. These characteristics have the potential to shape learning experiences in ways that may not be captured in studies examining AI as a general category. A more focused investigation of ChatGPT is therefore needed to better understand its particular role in supporting language learning, learner autonomy, and motivation among learners of Arabic as a foreign language.

[–>[–>

Taken together, the existing literature suggests that the educational potential of ChatGPT in Arabic language learning remains only partially understood. While research has documented the growing use of AI in language education, less attention has been paid to how learners employ such technologies as part of their self-directed learning practices, particularly in the context of Arabic as a foreign language. Furthermore, little is known about how learners evaluate the role of ChatGPT in supporting their language development, motivation, and autonomy. By addressing these issues, the present study seeks to contribute to a more nuanced understanding of how generative AI is shaping independent language learning and to provide insights that may inform the effective integration of AI-supported learning within Arabic language education.

[–>[–>

2.3 The current study

[–>[–>

Building on the gaps identified in the literature, the present study examines how non-native learners of Arabic use ChatGPT as part of their language-learning practices beyond formal classroom instruction. Although recent research has documented the growing role of artificial intelligence in language education, relatively little is known about how learners of Arabic employ ChatGPT to support self-directed learning or how they perceive its contribution to their language development. Guided by the theoretical perspectives of self-directed learning and learner autonomy, this study seeks to explore learners' experiences with ChatGPT and to better understand the role the technology plays in supporting independent language learning. Rather than focusing solely on learning outcomes, it considers how learners engage with the technology, the purposes for which they use it, and the role it plays in supporting their learning goals. Particular attention is given to how ChatGPT is used to develop different aspects of Arabic proficiency, including reading comprehension, writing, vocabulary knowledge, and grammatical competence.

[–>[–>

The current study also investigates learners' perceptions of ChatGPT as a language-learning resource. While recent discussions of generative AI have highlighted its potential educational benefits, learners' experiences are likely to be more complex and nuanced. For this reason, this study explores not only the perceived advantages of using ChatGPT but also the challenges and limitations learners encounter. In doing so, it seeks to develop a more balanced understanding of the role that AI-powered technologies can play in language learning.

[–>[–>

More specifically, the study addresses the following research questions:

[–>[–>

[–>
• [–>

  What self-directed learning practices do non-native Arabic learners use with ChatGPT?

  [–>

[–>

• [–>

  How do non-native Arabic learners perceive the impact of ChatGPT on their Arabic language development?

  [–>

[–>

• [–>

  How does ChatGPT influence non-native Arabic learners' motivation, autonomy, and literacy skills?

  [–>

[–>

By focusing on learners of Arabic as a foreign language within a Saudi university context, the study contributes to an area that remains comparatively underexplored in the literature on AI-assisted language learning. Examining learners' experiences can provide valuable insights into how generative AI technologies are being incorporated into independent language-learning practices and how learners make use of these tools to support their development. More broadly, the findings may contribute to ongoing discussions about the relationship between AI, learner autonomy, and self-directed language learning. They may also offer practical insights for educators, curriculum designers, and policymakers seeking to integrate AI technologies into Arabic language education in ways that support learner independence while maintaining the central importance of effective pedagogy and meaningful human interaction.

[–>

3 Methodology

[–>

3.1 Research design

[–>[–>

This study employed a qualitative research approach using an exploratory case study design to examine how non-native learners of Arabic use ChatGPT as a self-directed learning tool at the first author's university in Saudi Arabia. A case study approach was considered appropriate because it enables researchers to investigate a contemporary phenomenon within its real-life context and to gain a detailed understanding of participants' experiences and perspectives ([–>Yin, 2018).

[–>

The use of generative AI tools such as ChatGPT in Arabic language learning is still relatively new, and little is known about how learners use these tools independently outside formal classroom settings. For this reason, an exploratory design was selected. Rather than testing predetermined hypotheses, the study sought to understand participants' experiences, perceptions, and learning practices and to identify aspects of ChatGPT use that may not yet be well documented in the literature.

[–>[–>

The study was guided by an interpretive perspective, with the assumption that learners construct their own understanding of educational technologies through personal experience and interaction. Accordingly, the focus was on how participants made sense of ChatGPT and how they integrated it into their Arabic language learning practices.

[–>[–>

The interview data were analyzed inductively, allowing themes to emerge from participants' accounts rather than being organized according to predefined categories. Although the research questions provided a general direction for the analysis, the researchers remained open to identifying unexpected patterns and insights that arose from the data.

[–>[–>

As with most qualitative case studies, the purpose of this research was not to produce statistically generalizable findings. Instead, the study aimed to develop a rich understanding of learners' experiences within a particular educational context. By providing detailed descriptions of the participants and research setting, readers can determine the extent to which the findings may be relevant to similar contexts.

[–>[–>

3.2 Participants

[–>[–>

The study involved six non-native learners of Arabic from diverse linguistic and cultural backgrounds, including the Philippines, Indonesia, India, Malaysia, Uganda, and Pakistan. Participants were enrolled in different academic programs at the university, including Sharia, Theology and Da'wah, Arabic Language and Humanities Studies, and Law and Economics.

[–>[–>

Participants were selected through purposive sampling because the study required individuals who had direct experience using ChatGPT for Arabic language learning. To be included in the study, participants had to meet three criteria. First, they had to be regular users of ChatGPT for at least 1 month prior to the study. Second, they needed to have used ChatGPT specifically to support their Arabic language learning. Third, they had to have completed at least 4 years of study at the university. These criteria were established to ensure that participants had sufficient experience with both Arabic language learning and the use of ChatGPT to provide informed and reflective accounts.

[–>[–>

The sample consisted of four undergraduate students and two postgraduate students aged between 24 and 29 years (M  = 26.3). تم أيضًا جمع معلومات إضافية، بما في ذلك مستوى إتقان اللغة العربية، ومدة الدراسة باللغة العربية، وتكرار استخدام ChatGPT، لتوفير فهم أكمل لخلفيات المشاركين (انظر[–>[–>Table 1).

[–>

The sample size was relatively small; however, this was consistent with the exploratory nature of the study and with the goals of qualitative case study research, where depth of understanding is prioritized over broad representation. Data collection continued until thematic saturation was reached. By the sixth interview, participants were describing similar experiences and no substantially new themes were emerging. At that point, the researchers concluded that sufficient data had been collected to address the research questions.

[–>[–>

Although the findings offer valuable insights into learners' experiences, they should be interpreted within the context of a small group of participants from a single institution. The intention is therefore not to claim broad generalizability but to provide a detailed understanding of a particular group of Arabic learners and their engagement with ChatGPT.

[–>[–>

3.3 Instrument and data collection

[–>[–>

Data were collected through semi-structured interviews. This method was chosen because it provides a balance between structure and flexibility, allowing participants to share their experiences in detail while giving the interviewer opportunities to explore issues that emerged during the conversation.

[–>[–>

The interview guide was developed after reviewing literature related to self-directed learning, language learning technologies, artificial intelligence in education, and second-language acquisition. The questions were designed to encourage participants to reflect on how they used ChatGPT, the kinds of learning activities they engaged in, the benefits they perceived, and any challenges they encountered.

[–>[–>

Before data collection began, the interview protocol was reviewed by a scholar in applied linguistics to assess the clarity and relevance of the questions. Minor revisions were made based on the feedback received. A pilot interview was then conducted with a student who met the study criteria but did not participate in the final sample. This helped ensure that the questions were understandable and capable of eliciting detailed responses.

[–>[–>

The interviews were conducted face-to-face between March and April 2025. All interviews were carried out in Arabic to allow participants to express themselves comfortably and naturally. Each interview lasted approximately 30 min and was audio-recorded with participants' permission.

[–>[–>

Following the interviews, the recordings were transcribed verbatim and translated into English. Translation was undertaken by a proficient bilingual translator. To ensure accuracy, the translated transcripts were compared with the original Arabic recordings, and selected excerpts were back-translated to check for consistency in meaning. Any discrepancies were discussed and resolved during the review process.

[–>[–>

To maintain consistency across interviews, all interviews were conducted by the second author. Throughout the data collection process, the researchers kept field notes and reflective memos to document observations, emerging ideas, and potential sources of researcher influence on the research process.

[–>[–>

3.4 Data analysis

[–>[–>

The interview data were analyzed using inductive thematic analysis ([–>Braun and Clarke, 2022;[–>[–>Lochmiller, 2021). تم اختيار هذا النهج لأنه يسمح للموضوعات بالظهور من تجارب المشاركين بدلاً من فرض فئات محددة مسبقًا على البيانات.

[–>

Analysis began with repeated readings of the transcripts to develop familiarity with the data. During this stage, the researchers noted initial impressions, recurring ideas, and potentially meaningful patterns. Listening to interview recordings alongside the transcripts helped ensure that important contextual details were not overlooked.

[–>[–>

The next stage involved open coding. Segments of text that appeared relevant to the research questions were identified and assigned descriptive codes. Coding was conducted inductively, meaning that codes were derived from the data itself rather than from an existing theoretical framework.

[–>[–>

Once the initial coding was complete, related codes were grouped together to form broader categories. The researchers then reviewed these categories across all interviews to identify patterns that captured shared experiences among participants. As the analysis progressed, some categories were combined, revised, or renamed to improve clarity and coherence.

[–>[–>

The emerging themes were continuously compared against the original data to ensure that they accurately reflected participants' accounts. Particular attention was given to similarities and differences across participants, allowing both common patterns and individual perspectives to be identified.

[–>[–>

The analysis was conducted manually due to the relatively small size of the dataset. Working directly with the transcripts enabled the researchers to remain closely engaged with the data throughout the analytical process.

[–>[–>

To enhance the credibility of the findings, both researchers independently reviewed the coding and thematic structure. The researchers then met regularly to discuss their interpretations and resolve any differences in coding. Through these discussions, a shared understanding of the data and final themes was reached.

[–>[–>

3.5 Ethical considerations

[–>[–>

Ethical approval for this study was granted by the Standing Committee for Research Ethics at a university affiliated with one of the researchers (Reference No. H-1445-5-26) before the commencement of data collection. All participants received information about the purpose of the study, the interview procedures, and their rights as research participants. Participation was entirely voluntary, and written informed consent was obtained from all participants prior to their involvement in the study. Participants were also informed that they could withdraw from the study at any stage without penalty.

[–>[–>

To ensure confidentiality, pseudonyms were assigned to all participants, and any identifying information was removed from the interview transcripts and related records. Audio recordings, transcripts, and field notes were stored on an encrypted, password-protected device accessible only to the researchers. All data were treated as confidential and used solely for the purposes of this study. No information that could identify individual participants is reported in the findings.

[–>[–>

3.6 Trustworthiness

[–>[–>

Several strategies were employed to enhance the trustworthiness of the study.

[–>[–>

To strengthen credibility, the researchers conducted in-depth interviews and used follow-up questions to encourage participants to elaborate on their experiences. Throughout the analysis, the interview recordings and transcripts were revisited repeatedly to ensure that interpretations remained closely connected to participants' accounts. Participants were also invited to review summaries of their interviews and the researchers' preliminary interpretations. Their feedback confirmed that the summaries accurately reflected their experiences and perspectives.

[–>[–>

The researchers also engaged in ongoing discussions about coding decisions and theme development during the analytical process. In addition, an experienced qualitative researcher who was not directly involved in the study reviewed the developing analysis and provided feedback on the interpretation of the data. These discussions helped challenge assumptions and strengthen the overall rigor of the analysis.

[–>[–>

To support dependability, the researchers maintained a detailed record of important decisions made during data collection and analysis. This documentation included interview notes, coding records, analytical memos, and reflections on the research process.

[–>[–>

Confirmability was enhanced through reflexive practice. The researchers kept reflective notes throughout the study and regularly considered how their own experiences and assumptions might influence the interpretation of participants' responses. Independent review of the coding and subsequent discussion also helped ensure that the findings remained grounded in the data.

[–>[–>

Transferability was addressed by providing detailed descriptions of the participants, the research setting, and the procedures followed throughout the study. These descriptions enable readers to evaluate the relevance of the findings to other educational contexts.

[–>

4 Findings

This study explored how non-native learners of Arabic at a Saudi university use ChatGPT to support their self-directed learning. Analysis of the interview data revealed two overarching themes. The first theme relates to the use of ChatGPT for improving Arabic language skills, while the second concerns its role in enhancing motivation for self-directed learning. Within these two themes, six sub-themes emerged from participants' experiences. [–>Table 2 يقدم نظرة عامة على المواضيع والمواضيع الفرعية التي تم تحديدها من خلال التحليل.

بشكل عام، نظر المشاركون إلى ChatGPT كمورد قيم يدعم تطوير اللغة والتعلم المستقل. ووصفوا استخدام الأداة لتحسين المفردات والقواعد واستيعاب القراءة، مع الاعتماد عليها أيضًا للحصول على مساعدة فورية عند مواجهة صعوبات التعلم. وإلى جانب فوائدها اللغوية، سلط المشاركون الضوء على التأثير المحفز للاستجابات الفورية، والملاحظات التصحيحية، والوصول المستمر إلى الدعم. يبدو أن هذه الميزات تشجع المتعلمين على تحمل مسؤولية أكبر في تعلمهم والانخراط بشكل أكثر نشاطًا في اللغة العربية خارج الفصل الدراسي. تتم مناقشة هذه المواضيع والمواضيع الفرعية بالتفصيل أدناه مع اقتباسات داعمة من المشاركين.

[–>

4.1 Theme 1: using ChatGPT to improve Arabic language learning skills

[–>[–>

One of the most prominent themes emerging from the interviews was the contribution of ChatGPT to the development of Arabic language skills. All participants described incorporating the tool into their learning routines in different ways. Whether they were learning new vocabulary, clarifying grammatical rules, or trying to understand challenging texts, ChatGPT was frequently used as a source of explanation and guidance. Participants often referred to the convenience of accessing support whenever they needed it, allowing them to continue learning outside formal classroom settings.

[–>[–>[–>

4.1.1 Sub-theme 1: vocabulary improvement

[–>[–>

Vocabulary development was among the most commonly cited benefits of using ChatGPT. Participants explained that they frequently consulted the tool when they encountered unfamiliar words, wanted to understand different meanings, or needed examples of how words were used in context. Many felt that interacting with ChatGPT exposed them to a wider range of vocabulary than they would normally encounter during regular classroom activities. For example, P2 described how conversational exchanges with ChatGPT encouraged vocabulary growth:

[–>[–>

[–>

By using ChatGPT to create vocabulary exercises through daily conversations with ChatGPT, students can naturally expand their vocabulary through the repeated use of words in different contexts (P2).

[–>

[–>

Similarly, P6 explained that seeing words used in multiple contexts made them easier to remember:

[–>[–>

[–>

I like to ask ChatGPT about words that I do not understand well. I can get the meanings of the words, along with how they are used in different sentences, and this helps me remember them better (P6).

[–>

[–>

Several participants suggested that vocabulary learning often occurred naturally during their interactions with the tool. As P4 noted:

[–>[–>

[–>

Whenever I interact with ChatGPT, I come across new words and expressions that I have not learned before, and this helps me increase my vocabulary (P4).

[–>

[–>

Vocabulary learning was also closely connected to broader language development. P3 highlighted this relationship when explaining:

[–>[–>

[–>

Through interacting with ChatGPT, I learned more about grammar rules and sentence structure in Arabic. It's not just about learning vocabulary, but also about learning how to organize words correctly in sentences (P3).

[–>

[–>

Overall, participants viewed ChatGPT as a useful resource for expanding their vocabulary through exposure to meanings, examples, and contextualized language use.

[–>[–>[–>

4.1.2 Sub-theme 2: improving grammatical knowledge

[–>[–>

Grammar learning emerged as another important area in which participants found ChatGPT helpful. Many participants explained that they used the tool whenever they encountered difficulties with Arabic grammatical rules, particularly concepts that they considered complex or confusing. Rather than simply providing answers, ChatGPT was valued for its ability to explain rules and demonstrate how they function in authentic language use. For instance, P1 emphasized the usefulness of examples in understanding grammatical concepts:

[–>[–>

[–>

Students can request examples of how different grammatical rules are used in sentences, which helps in understanding the practical application of the rules (P1).

[–>

[–>

Likewise, P4 described how the tool supported the learning of more challenging grammatical structures:

[–>[–>

[–>

In fact, I use ChatGPT especially when I want to learn how to conjugate irregular verbs in correct sentences (P4).

[–>

[–>

Participants also appreciated the role of ChatGPT in identifying and correcting mistakes. For P5, grammar learning often occurred through receiving explanations about language errors:

[–>[–>

[–>

When I make mistakes in writing or speaking, ChatGPT helps me identify and correct those mistakes, which improves my understanding of grammar (P5).

[–>

[–>

Similarly, P6 described how detailed explanations contributed to a clearer understanding of sentence construction:

[–>[–>

[–>

I can ask ChatGPT to explain the grammatical structure of a sentence step by step, and this makes it easier to understand how the sentence is formed (P6).

[–>

[–>

These accounts suggest that participants viewed ChatGPT as more than a reference tool. It was frequently used as a learning companion that helped them explore grammatical concepts, verify their understanding, and learn from their mistakes. Such practices reflect learners' efforts to take an active role in addressing their own learning needs.

[–>[–>[–>

4.1.3 Sub-theme 3: improving Reading comprehension

[–>[–>

Reading comprehension was another area in which participants perceived considerable benefits from using ChatGPT. Learners described turning to the tool when they encountered difficult words, complex sentence structures, or passages that required deeper interpretation. For many, the ability to ask questions about a text and receive immediate clarification made reading Arabic materials more manageable and enjoyable. For example, P2 explained how ChatGPT could assist learners in understanding texts:

[–>[–>

[–>

Students can submit texts in Arabic (such as excerpts from books or articles) to ChatGPT and request an explanation of difficult words or an interpretation of complex sentences (P2).

[–>

[–>

Participants also highlighted the role of ChatGPT in helping them understand literary and cultural meanings. P4 explained:

[–>[–>

[–>

I can interact with literary texts, so non-native Arabic speakers can study Arabic literature or short stories and ask ChatGPT to explain the cultural or literary messages behind the texts (P4).

[–>

[–>

P5 provided a similar example:

[–>[–>

[–>

If a non-native Arabic speaker presents a poem like verses from the poem ‘If the people one day desire life', they can ask ChatGPT about the meaning of difficult words or inquire about the literary and cultural message behind these verses (P5).

[–>

[–>

For P6, contextual explanations played a particularly important role in comprehension:

[–>[–>

[–>

When reading novels or articles, I can ask ChatGPT about the meaning of a sentence in its context, which helps me understand the text better (P6).

[–>

[–>

Participants' accounts suggest that ChatGPT supported reading comprehension on multiple levels. In addition to helping learners understand vocabulary and sentence structures, it also enabled them to engage with broader cultural and literary dimensions of Arabic texts. This allowed learners to move beyond surface-level understanding and develop deeper interpretations of what they were reading.

[–>[–>

4.2 Theme 2: using ChatGPT to boost students' motivation for self-directed learning

[–>[–>

Alongside its contribution to language development, ChatGPT was also viewed as an important source of motivation for self-directed learning. Participants consistently described the tool as something that encouraged them to study independently and engage more actively with Arabic outside formal learning environments. The accessibility of ChatGPT, together with its ability to provide immediate support, appeared to make learners feel more confident in taking control of their learning.

[–>[–>

Three interconnected factors emerged from the interviews: the personal and interactive nature of communication with ChatGPT, the provision of immediate feedback, and the encouragement of greater independence in learning.

[–>[–>[–>

4.2.1 Sub-theme 1: personal interaction and instant responses

[–>[–>

A recurring idea across participant responses was the value of receiving immediate answers to questions. Participants explained that instant access to explanations helped them overcome learning obstacles without interrupting their study sessions. Rather than postponing questions until they could consult a teacher, learners were able to obtain clarification immediately and continue learning. P3 illustrated this experience with the example of learning grammar:

[–>[–>

[–>

If a non-native Arabic speaker studies Arabic grammar and encounters difficulty understanding the rule of tanween (nunation), they can ask an immediate question to ChatGPT about how to use tanween in different sentences. By receiving an instant answer, the student feels confident and accomplished, which boosts their motivation to continue learning the language rules (P3).

[–>

[–>

Similarly, P5 described how immediate responses encouraged continued engagement with learning:

[–>[–>

[–>

Whenever I have a question about Arabic language or culture, I can ask ChatGPT and receive an answer immediately. This makes learning easier and encourages me to keep studying (P5).

[–>

[–>

Beyond the speed of responses, participants frequently referred to the interactive nature of their conversations with ChatGPT. P3 explained:

[–>[–>

[–>

ChatGPT provides non-native Arabic speakers with a sense of continuous interaction, which enhances their feeling of engagement and progress in learning the language (P3).

[–>

[–>

A similar sentiment was expressed by P2:

[–>[–>

[–>

The interaction feels similar to having someone available all the time to answer questions and explain language concepts whenever needed (P2).

[–>

[–>

For many learners, this constant availability reduced frustration and created a sense of support throughout the learning process. The combination of rapid responses and interactive communication appeared to play an important role in sustaining motivation and encouraging continued engagement with Arabic learning activities.

[–>[–>[–>

4.2.2 Sub-theme 2: immediate feedback

[–>[–>

Another important motivational factor identified by participants was the availability of immediate feedback. Learners frequently described how ChatGPT could quickly identify errors and provide corrections, helping them recognize weaknesses and make improvements promptly. For instance, P1 explained:

[–>[–>

[–>

When students write texts or sentences in Arabic, they can receive immediate corrections from ChatGPT. These corrections give students a sense of continuous progress and improvement, motivating them to continue learning (P1).

[–>

[–>

Similarly, P4 highlighted the encouraging nature of corrective feedback:

[–>[–>

[–>

When ChatGPT shows me my mistakes and explains how to correct them, I feel that I am improving, and this encourages me to continue practising Arabic (P4).

[–>

[–>

For P2, feedback was closely connected to confidence building:

[–>[–>

[–>

The feedback helps students understand what they did wrong and how to improve. This makes them feel more confident in using Arabic (P2).

[–>

[–>

Likewise, P6 appreciated that the corrections were accompanied by explanations:

[–>[–>

[–>

ChatGPT not only corrects errors but also explains the reasons behind the corrections, which helps me avoid making the same mistake again (P6).

[–>

[–>

Several participants also noted that receiving feedback from ChatGPT felt less intimidating than receiving feedback in other situations. The absence of judgment appeared to make learners more willing to experiment with the language and learn from their mistakes. As a result, feedback was perceived not only as a means of improving language performance but also as an important source of encouragement and confidence.

[–>[–>[–>

4.2.3 Sub-theme 3: encouraging Independence and self-directed learning

[–>[–>

The final sub-theme relates to learner autonomy. Participants consistently described ChatGPT as a tool that enabled them to become more independent in their studies. By providing explanations, examples, and guidance whenever needed, the tool reduced learners' dependence on teachers and allowed them to address difficulties on their own. As explained by P3:

[–>[–>

[–>

ChatGPT enhances students' independence and encourages them to learn without fully relying on traditional teachers. Students feel that they are capable of learning Arabic on their own, which boosts their confidence (P3).

[–>

[–>

Similarly, P4 reflected on how the tool encouraged exploration and initiative:

[–>[–>

[–>

From my experiences with using ChatGPT, by interacting with the tool, I can discover new vocabulary and topics on my own, which enhances my motivation for exploration and self-directed learning (P4).

[–>

[–>

The flexibility offered by ChatGPT was also highlighted by P2:

[–>[–>

[–>

I can study whenever I want and focus on the language areas that I personally need to improve. This makes me more responsible for my own learning (P2).

[–>

[–>

Likewise, P6 described using the tool as a means of solving problems independently:

[–>[–>

[–>

When I face difficulties in grammar or vocabulary, I can use ChatGPT to find explanations by myself instead of waiting for help from others (P6).

[–>

[–>

Participants further explained that this sense of independence encouraged them to engage with Arabic more frequently outside formal educational settings. Having continuous access to support enabled them to set their own learning goals, focus on areas of weakness, and learn at a pace that suited their individual needs. As they became more confident in managing their learning, many reported taking greater initiative in exploring new language topics and monitoring their own progress.

[–>[–>

Overall, the findings indicate that ChatGPT contributed to a stronger sense of learner autonomy. By giving students greater control over when, how, and what they learned, the tool appeared to support the development of self-directed learning habits and increased confidence in studying Arabic independently.

[–>

5 Discussion

This study set out to explore how non-native learners of Arabic perceive the use of ChatGPT as a tool for autonomous language learning. Overall, participants described ChatGPT as a helpful and accessible resource that supported different aspects of their learning, particularly vocabulary development, grammar understanding, reading comprehension, motivation, and independent learning. Although these findings broadly reflect trends reported in previous studies on AI-assisted language learning, the present study contributes a perspective that has received relatively limited attention: the experience of learners studying Arabic as a foreign language.

Vocabulary development emerged as one of the most noticeable benefits reported by participants. Many learners explained that interacting with ChatGPT exposed them to new words in meaningful contexts rather than in isolated lists. This appears to be particularly important in Arabic, where understanding a word often requires recognizing its relationship to a root and its various derived forms. The participants' experiences therefore suggest that the value of ChatGPT may extend beyond introducing new vocabulary items; it may also help learners make sense of the relationships between words and their patterns of use. This interpretation is consistent with earlier research highlighting the importance of contextualized exposure for vocabulary learning ([–>Boudouaia et al., 2024;[–>[–>Kohnke et al., 2023). وفي الوقت نفسه، تشير النتائج إلى أن المتعلمين لا يحفظون المفردات فحسب، بل يتفاعلون بنشاط مع اللغة بطرق تساعدهم على بناء روابط بين الشكل والمعنى.

النتائج المتعلقة بالقواعد جديرة بالملاحظة أيضًا. غالبًا ما ينظر المتعلمون غير الأصليين إلى قواعد اللغة العربية باعتبارها واحدة من أكثر مكونات تعلم اللغة تحديًا بسبب نظامها التصريفى المعقد واعتمادها على علامات الحالة النحوية. أكد المشاركون مرارًا وتكرارًا على أن ChatGPT جعل المفاهيم النحوية الصعبة أكثر قابلية للفهم من خلال تقديم تفسيرات وأمثلة فورية. في حين أن هذا يدعم الحجج السابقة المتعلقة بالقيمة التربوية للتفسيرات التي يولدها الذكاء الاصطناعي ([–>[–>AlAwfi, 2024)، فهو يثير أيضًا سؤالًا مهمًا. هل يقوم المتعلمون بالفعل بتطوير فهم أعمق للهياكل النحوية، أم أنهم ببساطة يجدون التفسيرات أسهل في المتابعة؟ التمييز مهم لأن الفهم المتصور لا يترجم بالضرورة إلى التعلم على المدى الطويل. إن بساطة وإمكانية الوصول إلى تفسيرات ChatGPT قد تساهم في ثقة المتعلمين، ولكن الثقة وحدها لا يمكن اعتبارها دليلاً على الإتقان النحوي. ويرتبط هذا بشكل خاص بالتعلم المستقل لأنه يقلل من اعتماد المتعلمين على تدخل المعلم ويسمح لهم بمعالجة الصعوبات اللغوية بشكل مستقل. لذلك تشير النتائج إلى أن مساهمة ChatGPT تمتد إلى ما هو أبعد من دعم القواعد النحوية نفسها لتسهيل سيطرة المتعلم بشكل أكبر على عملية التعلم. لذلك ستحتاج الأبحاث المستقبلية إلى فحص ما إذا كانت هذه التصورات الإيجابية تنعكس في تحسينات قابلة للقياس في أداء اللغة.

ويمكن ملاحظة توتر مماثل في النتائج المتعلقة بفهم القراءة. وصف المشاركون ChatGPT بأنه وسيلة دعم مفيدة عند العمل مع النصوص العربية، خاصة عندما تقاطع المفردات غير المألوفة أو التعبيرات المعقدة فهمهم. وبدلاً من التخلي عن النصوص الصعبة، تمكن المتعلمون من طرح الأسئلة وطلب التوضيح وتلقي المساعدة الفورية. يبدو أن هذا يقلل من أحد العوائق الرئيسية التي يواجهها العديد من متعلمي اللغة: الإحباط الذي يصاحب غالبًا قراءة المواد الأصلية. ومن المثير للاهتمام أن بعض المشاركين أفادوا أيضًا باستخدام ChatGPT لاستكشاف المعاني الثقافية والأدبية المضمنة في النصوص. يشير هذا إلى أن الأداة قد تدعم مستوى أعمق من المشاركة مقارنة بالفهم البسيط. ومع ذلك، هناك أيضًا جانب سلبي محتمل. إذا اعتاد المتعلمون على الاعتماد على التفسيرات التي يولدها الذكاء الاصطناعي كلما واجهوا صعوبة، فقد يقضون وقتًا أقل في تطوير المهارات التحليلية والاستنتاجية التي تتطلبها القراءة الفعالة عادة. وبهذا المعنى، يمكن أن يعمل ChatGPT إما كسقالة تعزز التعلم أو كاختصار يحد من المشاركة الفكرية، اعتمادًا على كيفية استخدامه.

ظهر الدافع كموضوع بارز آخر طوال فترة الدراسة. أشار المشاركون بشكل متكرر إلى سهولة توفر ChatGPT، مشيرين إلى أنه يمكنهم استخدامه وقتما يريدون دون الحاجة إلى انتظار معلم أو شريك لغوي. يبدو أن إمكانية الوصول المستمرة هذه قد شجعت على المزيد من المشاركة المتكررة في تعلم اللغة العربية. تعتبر مثل هذه النتيجة ذات أهمية خاصة لأن فرص ممارسة اللغة الأصيلة غالبًا ما تكون محدودة بالنسبة لمتعلمي اللغة العربية غير الأصليين. على عكس التدريس في الفصول الدراسية، والذي يقتصر على الزمان والمكان، يوفر ChatGPT بيئة تعليمية متاحة باستمرار. ومع ذلك، لا ينبغي تفسير زيادة المشاركة تلقائيًا على أنها زيادة في التعلم. قد يقضي المتعلمون وقتًا أطول في التفاعل مع اللغة لمجرد أن التجربة ممتعة ومريحة. وفي حين أن هذا مفيد بلا شك، فإن الدافع وحده لا يمكن أن يضمن تحقيق تقدم ملموس. وبالتالي فإن التحدي الذي يواجه المعلمين هو الاستفادة من المزايا التحفيزية لأدوات الذكاء الاصطناعي مع ضمان استمرار تركيز المتعلمين على أهداف تعليمية محددة بوضوح.

إن الأهمية التي يعلقها المشاركون على التعليقات الفورية تسلط الضوء بشكل أكبر على جاذبية ChatGPT كمورد تعليمي. قدّر العديد من المتعلمين القدرة على تلقي التصحيحات والتفسيرات في اللحظة المحددة التي واجهوا فيها مشكلة. تخلق هذه السرعة تجربة تعليمية تختلف بشكل كبير عن البيئات التعليمية التقليدية، حيث تتأخر ردود الفعل في كثير من الأحيان. وفي الوقت نفسه، لا ينبغي افتراض أن جودة التعليقات الفورية معادلة لجودة تعليقات الخبراء. على عكس المعلمين البشريين، قد تتجاهل أنظمة الذكاء الاصطناعي العوامل السياقية، أو تفشل في التعرف على الصعوبات الدقيقة التي يواجهها المتعلم، أو تقدم في بعض الأحيان استجابات غير دقيقة. وبالتالي، ينبغي النظر إلى سرعة ردود الفعل باعتبارها إحدى نقاط قوة ChatGPT، في حين تظل موثوقيتها أمرًا يحتاج المتعلمون إلى تقييمه بشكل نقدي.

وللنتائج أيضًا آثار مهمة على أصول تدريس اللغة العربية. بدلاً من النظر إلى ChatGPT كبديل للمعلمين، تشير النتائج إلى أنه قد يكون أكثر فعالية عند استخدامه جنبًا إلى جنب مع أشكال التدريس التقليدية. يمكن للمدرسين تشجيع المتعلمين على استخدام ChatGPT لمزيد من التدريب والمراجعة والاستكشاف مع مساعدتهم أيضًا على تقييم جودة المعلومات التي يتلقونها. وبهذه الطريقة، يصبح الذكاء الاصطناعي أداة تعمل على توسيع فرص التعلم خارج الفصل الدراسي بدلاً من أن تحل محل الممارسات التربوية الراسخة. يسلط الحضور المتزايد لتقنيات الذكاء الاصطناعي في التعليم الضوء أيضًا على أهمية تطوير المعرفة الرقمية المهمة لدى المتعلمين. لا يحتاج الطلاب إلى معرفة كيفية استخدام أدوات الذكاء الاصطناعي فحسب، بل يحتاجون أيضًا إلى معرفة كيفية طرح المعلومات التي ينشئونها والتحقق منها وتقييمها.

وعلى الرغم من الآراء الإيجابية التي أبداها المشاركون، تشير الدراسة أيضا إلى عدة تحديات لا ينبغي إغفالها. وتظل المخاوف بشأن دقة المعلومات التي يولدها الذكاء الاصطناعي ذات أهمية خاصة في حالة اللغة العربية، حيث تكون تقنيات اللغة بشكل عام أقل تطوراً من اللغة الإنجليزية. يمكن أن تؤدي الأخطاء في التفسيرات النحوية أو الترجمات أو التفسيرات الثقافية إلى تضليل المتعلمين إذا تم قبول المخرجات دون التحقق منها. هناك مصدر قلق آخر يتعلق بالإفراط في الاعتماد على دعم الذكاء الاصطناعي. كلما زاد اعتماد المتعلمين على الإجابات الفورية، قل احتمال انخراطهم في حل المشكلات بشكل مستقل أو التفكير النقدي. تشير هذه المشكلات إلى أن الدمج الناجح لـ ChatGPT في تعلم اللغة يتطلب نهجًا متوازنًا يجمع بين مزايا الذكاء الاصطناعي والتوجيه التربوي المناسب ووعي المتعلم.

في الختام، تشير النتائج إلى أن ChatGPT لديه إمكانات كبيرة لدعم تعلم اللغة العربية بشكل مستقل. ويبدو أن قيمته لا تكمن في المعلومات التي يقدمها فحسب، بل أيضًا في قدرته على خلق فرص للممارسة المستمرة والاستكشاف والمشاركة. ومع ذلك، ينبغي تفسير الفوائد التي أبلغ عنها المشاركون في ضوء اعتماد الدراسة على التصورات المبلغ عنها ذاتيا. في النهاية، تشير النتائج إلى أن ChatGPT يكون أكثر فعالية عندما يتم التعامل معه باعتباره رفيقًا للتعلم وليس كمصدر موثوق للمعرفة. إذا تم استخدامه بشكل مدروس ونقدي، فيمكنه إثراء تجربة تعلم اللغة مع استكمال التوجيهات المقدمة من المعلمين والتعليم الرسمي.

6 Conclusion

This study explored how non-native learners of Arabic experience the use of ChatGPT as a tool for autonomous learning. The participants generally viewed ChatGPT as a helpful resource that supported different aspects of their language learning, particularly vocabulary development, grammar learning, and reading comprehension. The ability to receive immediate explanations and feedback appeared to make learning more accessible and less dependent on teacher support, allowing learners to address difficulties whenever they arose. In this way, ChatGPT functioned not only as a source of information but also as a tool that supported greater learner autonomy and self-direction.

An important finding was the role ChatGPT played in encouraging learners to take greater responsibility for their own learning. Participants described using the tool not only to answer questions but also to explore language independently, practice at their own pace, and revisit areas they found challenging. For many learners, this flexibility seemed to increase both their engagement and their confidence. Rather than being limited to classroom activities, learning could continue whenever support was needed.

The study also sheds light on the growing role of AI in Arabic language education. Much of the existing research on AI-assisted language learning has focused on English and other widely taught languages. By examining learners of Arabic, this study provides evidence that similar benefits may be observed in a different linguistic context, despite the particular challenges that Arabic presents for non-native learners.

At the same time, the findings should not be taken to mean that ChatGPT can replace teachers or formal instruction. Participants valued the support provided by the tool, but effective language learning still depends on guidance, practice, and opportunities for meaningful interaction. ChatGPT appears to be most useful when viewed as a supplementary resource that works alongside established teaching practices rather than as an alternative to them.

Overall, the findings point to the potential of ChatGPT to support autonomous Arabic language learning. Its value lies not only in the information it provides but also in the opportunities it creates for learners to engage with the language more independently and consistently. As AI technologies become increasingly present in educational settings, understanding how they can be used in ways that genuinely support learning will become even more important.

7 Limitations and future research

The findings of this study should be considered in light of several limitations.

First, the study involved a relatively small number of participants. Although the sample provided rich insights into learners' experiences, it does not allow for broad generalizations about all non-native learners of Arabic. Different learner populations may use ChatGPT in different ways or report different experiences.

Second, all participants were drawn from a single institutional context. Learning environments vary considerably in terms of teaching approaches, technological resources, and learner backgrounds. As a result, the experiences reported in this study may reflect characteristics of the specific context in which the research was conducted.

Another limitation concerns the type of data collected. The study relied mainly on participants' own accounts of how ChatGPT influenced their learning. While these perceptions are valuable because they reveal how learners make sense of their experiences, they cannot confirm whether actual improvements in language proficiency occurred. What learners believe has helped them may not always correspond to measurable learning gains.

The study was also limited by the absence of additional data sources. Classroom observations, language assessments, learner journals, or analysis of students' language performance could have provided a more complete understanding of ChatGPT's impact. The use of multiple sources of evidence would have strengthened the study by allowing learners' perceptions to be compared with observable learning outcomes.

Building on these findings, future research could examine larger and more diverse groups of Arabic learners to determine whether similar patterns emerge across different contexts. Studies conducted across multiple institutions would also help identify how local educational conditions influence the use of AI tools for language learning.

There is also considerable scope for mixed-methods research that combines learners' experiences with objective measures of language development. Such studies could provide a clearer picture of the relationship between perceived benefits and actual learning outcomes. In addition, future research could compare learners who use ChatGPT with those who rely on more traditional learning resources in order to better understand the specific contribution of AI-assisted learning.

Finally, longitudinal research is needed to explore the long-term effects of ChatGPT use. The present study captures learners' experiences at a particular point in time, but language learning is a gradual process that unfolds over months or years. Future studies could investigate whether the motivation, autonomy, and perceived learning benefits reported here are sustained over time and whether they lead to measurable improvements in Arabic language proficiency.

References

• [–>[–>[–>AlAwfi[–>[–>I. M.[–> ([–>2024).[–>Teaching functional morphology to non-native Arabic learners: a proposed conceptualization.[–>Theory Pract. Lang. Stud.[–>14,[–>3515– “[–>3524. دوى:[–>10.17507/tpls.1411.20

• [–>[–>[–>Almelhes[–>[–>S.[–> ([–>2024).[–>Enhancing Arabic language acquisition: effective strategies for addressing non-native learners' challenges.[–>Educ. Sci.[–>14:[–>1116. دوى:[–>10.3390/educsci14101116

• [–>[–>[–>Alshehri[–>[–>F. A.[–>[–>[–>Alshehri[–>[–>F. A.[–>[–>[–>Qassem[–>[–>M.[–>[–>[–>Bedjaoui[–>[–>W.[–>[–>[–>Alfaisal[–>[–>A. A.[–> ([–>2025).[–>The dynamics of translation from and into Arabic in the Arab world: bibliometric analysis of the index Translationum UNESCO database (1979–2012).[–>Humanit. Soc. Sci. Commun.[–>12:[–>43. دوى:[–>10.1057/s41599-024-04225-5

• [–>[–>[–>Benson[–>[–>P.[–> ([–>2013).[–>Teaching and Researching Autonomy.[–>London:[–>Routledge.

• [–>[–>[–>Boudouaia[–>[–>A.[–>[–>[–>Mouas[–>[–>S.[–>[–>[–>Kouider[–>[–>B.[–> ([–>2024).[–>A study on ChatGPT-4 as an innovative approach to enhancing English as a foreign language writing learning.[–>J. Educ. Comput. Res.[–>62,[–>1509– “[–>1537. دوى:[–>10.1177/07356331241247465

• [–>[–>[–>Braun[–>[–>V.[–>[–>[–>Clarke[–>[–>V.[–> ([–>2022).[–>Thematic analysis: A practical guide.[–>QMiP Bulletin.[–>1,[–>46– “[–>50. دوى:[–>10.53841/bpsqmip.2022.1.33.46

• [–>[–>[–>El Zahraa[–>[–>F.[–> ([–>2025).[–>Leveraging artificial intelligence and digital technologies to enhance sociolinguistic competence and Arabic language Skills.[–>Proc. Int. Collab. Conf. Multidiscip. Sci.[–>2,[–>8– “[–>19. دوى:[–>10.70062/iccms.v2i1.65

• [–>[–>[–>Holec[–>[–>H.[–> ([–>1981).[–>Autonomy and Foreign Language Learning.[–>Oxford:[–>Pergamon Press.

• [–>[–>[–>Jiang[–>[–>H.[–>[–>[–>Cheng[–>[–>Y.[–>[–>[–>Yang[–>[–>J.[–>[–>[–>Gao[–>[–>S.[–> ([–>2022).[–>AI-powered chatbot communication with customers: dialogic interactions, satisfaction, engagement, and customer behavior.[–>Comput. Hum. Behav.[–>134:[–>107329. دوى:[–>10.1016/j.chb.2022.107329

• [–>[–>[–>Kasneci[–>[–>E.[–>[–>[–>Sessler[–>[–>K.[–>[–>[–>Küchemann[–>[–>S.[–>[–>[–>Bannert[–>[–>M.[–>[–>[–>Dementieva[–>[–>D.[–>[–>[–>Fischer[–>[–>F.[–>[–>et al[–>. ([–>2023).[–>ChatGPT for good? On opportunities and challenges of large language models for education.[–>Learn. Individ. Differ.[–>103:[–>102274. دوى:[–>10.1016/j.lindif.2023.102274

• [–>[–>[–>Knowles[–>[–>M. S.[–> ([–>1975).[–>Self-Directed Learning: A guide for Learners and Teachers.[–>New York:[–>Association Press.

• [–>[–>[–>Kohnke[–>[–>L.[–>[–>[–>Moorhouse[–>[–>B. L.[–>[–>[–>Zou[–>[–>D.[–> ([–>2023).[–>ChatGPT for language teaching and learning.[–>RELC J.[–>54,[–>537– “[–>550. دوى:[–>10.1177/00336882231162868

• [–>[–>[–>Li[–>[–>X.[–>[–>[–>Huang[–>[–>L.[–>[–>[–>Yao[–>[–>P.[–>[–>[–>Hyönä[–>[–>J.[–> ([–>2022).[–>Universal and specific reading mechanisms across different writing systems.[–>Nat. Rev. Psychol.[–>1,[–>133– “[–>144. دوى:[–>10.1038/s44159-022-00022-6

• [–>[–>[–>Lochmiller[–>[–>C. R.[–> ([–>2021).[–>Conducting thematic analysis with qualitative data.[–>Qual. Rep.[–>26,[–>2029– “[–>2044. دوى:[–>10.46743/2160-3715/2021.5008,[–>

• [–>[–>[–>Mohideen[–>[–>H. L. M.[–> ([–>2024).[–>Exploring the opportunities of implementing artificial intelligence (AI) technology for teaching Arabic to non-native speakers: a theoretical approach.[–>J. Dig. Learning and Dist. Edu[–>2,[–>760– “[–>767. دوى:[–>10.56778/jdlde.v2i9.225

• [–>[–>[–>Qpilat[–>[–>N.[–>[–>[–>Salam[–>[–>W.[–> ([–>2023).[–>Literary contexts for teaching non-Arabic learners.[–>Int. J. Humanit. Educ.[–>21,[–>75– “[–>89. دوى:[–>10.18848/2327-0063/CGP/v21i02/75-89

• [–>[–>[–>Ritonga[–>[–>M.[–>[–>[–>Mudinillah[–>[–>A.[–>[–>[–>Wasehudin[–>[–>W.[–>[–>[–>Julhadi[–>[–>J.[–>[–>[–>Amrina[–>[–>A.[–>[–>[–>Shidqi[–>[–>M. H.[–> ([–>2024).[–>The effect of technology on Arabic language learning in higher education.[–>J. Educ. Learn. (EduLearn)[–>18,[–>116– “[–>127. دوى:[–>10.11591/edulearn.v18i1.20867

• [–>[–>[–>Soliman[–>[–>R.[–>[–>[–>Khalil[–>[–>S.[–> ([–>2024).[–>The teaching of Arabic as a community language in the UK.[–>Int. J. Biling. Educ. Biling.[–>27,[–>1246– “[–>1257. دوى:[–>10.1080/13670050.2022.2063686

• [–>[–>[–>Yin[–>[–>R. K.[–> ([–>2018).[–>Case Study Research and Applications: Design and Methods.[–>Thousand Oaks:[–>SAGE.

• [–>[–>[–>ZainEldin[–>[–>H.[–>[–>[–>Gamel[–>[–>S. A.[–>[–>[–>Talaat[–>[–>F. M.[–>[–>[–>Aljohani[–>[–>M.[–>[–>[–>Baghdadi[–>[–>N. A.[–>[–>[–>Malki[–>[–>A.[–>[–>et al[–>. ([–>2024).[–>Silent no more: a comprehensive review of artificial intelligence, deep learning, and machine learning in facilitating deaf and mute communication.[–>Artif. Intell. Rev.[–>57:[–>188. دوى:[–>10.1007/s10462-024-10816-0