محركات اعتماد الذكاء الاصطناعي التوليدي في التعليم العالي من منظور نظرية السلوك المخطط Drivers of generative AI adoption in higher education through the lens of the Theory of Planned Behaviour

المجلة: Technology in Society، المجلد: 77
DOI: https://doi.org/10.1016/j.techsoc.2024.102521
تاريخ النشر: 2024-03-25

ملاحظة! هذه نسخة موازية من المقالة الأصلية. قد تختلف النسخة الموازية عن الأصل من حيث تنسيق الصفحات والمظهر.
يرجى ملاحظة! هذه نسخة إلكترونية مؤرشفة ذاتيًا من المقالة الأصلية. قد تختلف هذه النسخة عن الأصل في تنسيق الصفحات والتفاصيل الطباعية.
تشير إلى المصدر الأصلي:
استشهد بالنسخة النهائية:
إيفانوف، س.، سليمان، م.، تيوومي، أ.، الكثيري، ن. أ. و العلاوي، أ. ن. (2024) دوافع اعتماد الذكاء الاصطناعي التوليدي في التعليم العالي من خلال عدسة نظرية السلوك المخطط. التكنولوجيا في المجتمع، 77 (يونيو 2024)، 102521.https://doi.org/10.1016/j.techsoc.2024.102521
© 2024 المؤلفون. منشور بواسطة إلسفير المحدودة. هذه مقالة مفتوحة الوصول بموجب ترخيص CC BY (http://creativecommons.org/licenses/by/4.0/).

محركات اعتماد الذكاء الاصطناعي التوليدي في التعليم العالي من منظور نظرية السلوك المخطط

ستانيسلاف إيفانوف ، محمد سليمان ، أيرني تيوومي ، ناصر الحمير الكثيري ، الأمير ن. العلاوي جامعة فارنا للإدارة، 13A شارع أوبوريشت، 9000 فارنا، بلغاريا معهد زانغادور للبحوث، 9010 فارنا، بلغاريا قسم البحث والاستشارات، جامعة التكنولوجيا والعلوم التطبيقية، صلالة، عمان & كلية السياحة والفنادق، جامعة الفيوم، الفيوم، مصر جامعة هاكا-هيليا للعلوم التطبيقية، هلسنكي، فنلندا قسم إدارة الأعمال، كلية الاقتصاد وإدارة الأعمال، جامعة التكنولوجيا والعلوم التطبيقية-صلالة، صلالة، عمان جامعة التكنولوجيا والعلوم التطبيقية، إبراء، عمان

معلومات المقال

الكلمات المفتاحية:

الذكاء الاصطناعي التوليدي
نظرية السلوك المخطط
التعليم العالي

الملخص

استنادًا إلى نظرية السلوك المخطط (TPB)، تبحث هذه الدراسة في العلاقة بين الفوائد المدركة، والقوى، والضعف، والمخاطر لأدوات الذكاء الاصطناعي التوليدي (GenAI) والعوامل الأساسية لنموذج TPB (أي، الموقف، المعايير الذاتية، والتحكم السلوكي المدرك). كما تبحث الدراسة في الارتباط الهيكلي بين متغيرات TPB ونية استخدام أدوات GenAI، وكيف يمكن أن تؤثر الأخيرة على الاستخدام الفعلي لأدوات GenAI في التعليم العالي. تعتمد الورقة على نهج كمي، معتمدة على استبيان عبر الإنترنت يتم إدارته ذاتيًا بشكل مجهول لجمع بيانات أولية من 130 محاضرًا و168 طالبًا في مؤسسات التعليم العالي (HEIs) في عدة دول، وPLS-SEM لتحليل البيانات. تشير النتائج إلى أنه على الرغم من اختلاف تصورات المحاضرين والطلاب حول المخاطر والضعف لأدوات GenAI، فإن القوى والفوائد المدركة لتقنيات GenAI لها تأثير كبير وإيجابي على مواقفهم، والمعايير الذاتية، والتحكم السلوكي المدرك. تؤثر المتغيرات الأساسية لنموذج TPB بشكل إيجابي وكبير على نوايا المحاضرين والطلاب لاستخدام أدوات GenAI، والتي بدورها تؤثر بشكل كبير وإيجابي على اعتمادهم لمثل هذه الأدوات. تعزز هذه الورقة النظرية من خلال توضيح العوامل التي تشكل اعتماد تقنيات GenAI في HEIs. كما تقدم لأصحاب المصلحة مجموعة متنوعة من الآثار الإدارية والسياسية حول كيفية صياغة قواعد وأنظمة مناسبة للاستفادة من مزايا هذه الأدوات مع التخفيف من آثار عيوبها. كما تم توضيح القيود وفرص البحث المستقبلية.

1. المقدمة

بدأت الثورة في صناعة الخدمات، بما في ذلك القطاع التعليمي، عندما تقدمت التكنولوجيا بسرعة من حيث الذكاء والقوة بينما أصبحت أكثر إحكامًا وخفةً وبأسعار معقولة. يشمل ذلك كل من الأجهزة مثل تقنيات الخدمة الذاتية الذكية والبرمجيات والأنظمة مثل التعلم الآلي (ML) وأدوات الذكاء الاصطناعي التوليدي (GenAI) [1]. GenAI هو مصطلح يستخدم لوصف فئة من نماذج الذكاء الاصطناعي التي تنتج مخرجات جديدة ظاهريًا مثل النصوص، والصور، والفيديو، والموسيقى، أو أنواع أخرى من الوسائط. بينما كانت أساليب GenAI متاحة منذ فترة، فإن إطلاق ChatGPT أثار سيلًا
من المناقشات في وسائل الإعلام، والمنتديات عبر الإنترنت، والمجتمعات الأكاديمية [2-8]. نتيجة لذلك، أصبح الباحثون والممارسون مهتمين بشكل متزايد بتداعيات تطبيقات GenAI، خاصة تلك المعتمدة على نماذج اللغة الكبيرة (LLMs)، على التعلم البشري، وتوليد المعرفة، وطبيعة العمل في السنوات القادمة [4]. أشار إيفانوف وسليمان [9] إلى أنه على المدى الطويل، ستحدث الدردشة المعتمدة على LLM ثورة في البحث والتعليم. إذا تم اعتمادها بنجاح، يمكن استخدامها كمدرسين عبر الإنترنت، ومطوري مناهج، ومصححين، ومساهمين في المنشورات الأكاديمية. ستكون LLMs أيضًا ضرورية في إعادة التفكير في التعليم من تفاعلات “المعلم-الطالب” إلى “المعلم-الذكاء الاصطناعي-الطالب” في التعاون [9]، مما يحول
التركيز من المحاضرين إلى تطوير مهام وأنشطة جديدة مع تطبيقات GenAI. وهذا يبرر العدد المتزايد من المنشورات الحديثة حول مزايا وعيوب استخدام تقنيات GenAI، مثل ChatGPT، في البحث والتعليم (على سبيل المثال، المرجع [9-12]).
نظرًا للفرص الكبيرة لاستخدام GenAI، هناك حاجة ملحة لتحويل التركيز الأكاديمي من التأسف على انهيار التعليم والبحث إلى التفكير في كيفية استخدام الطلاب والباحثين لمثل هذه الأدوات [7]. وفقًا لمجاهد وآخرون [13]، فإن قدرة نماذج GenAI على توفير التعليم، وشرح المفاهيم الأساسية، وتوليد المعرفة قد تحدث ثورة في ممارسة التحكم في العمليات الإحصائية، والتعليم، والبحث. ومع ذلك، لا تزال هذه التقنيات في مراحلها الأولى من النشر وعرضة للإساءة وسوء الفهم. لذلك، هناك حاجة إلى تحليل تجريبي شامل لتقديم نظرة عامة وفهم عميق لإمكانات استخدام مثل هذه التطبيقات لأغراض التعليم والبحث. قدمت الدراسات السابقة رؤى قيمة حول استخدام أدوات GenAI في سياقات مختلفة بما في ذلك السياق التعليمي والبحثي من خلال عدسات نظريات مختلفة (على سبيل المثال، المرجع [9،12،14-16]). على سبيل المثال، باستخدام نظرية القبول والاستخدام الموحد للتكنولوجيا (UTAUT)، أظهر ستريزليكي والعربي [17] كيف أن التأثير الاجتماعي، وتوقع الجهد، وتوقع الأداء لها تأثير كبير على النية السلوكية. تأثرت سلوكيات استخدام ChatGPT الفعلية لطلاب الجامعات المصرية والبولندية بالنية السلوكية عند اعتبارها مع الظروف الميسرة. بحث جابوب وآخرون [18] كيف تأثرت إنجازات الطلاب الجامعيين بالأدوات والتطبيقات GenAI في ثلاث دول عربية: اليمن، والأردن، وعمان. أظهرت النتائج أن إنجازات الطلاب في HEIs العربية تأثرت بشكل إيجابي وكبير بأساليب وتطبيقات GenAI. بالإضافة إلى ذلك، أظهرت النتائج أن سلوك الطلاب عزز العلاقة بين أدوات GenAI وإنجازات المعرفة. استنادًا إلى نموذج UTAUT2 ومؤشر جاهزية التكنولوجيا، قيمت دراسة وانغ وزانغ [19] العناصر والسمات الشخصية التي دفعت الجيل Z لاعتماد التصميم المدعوم بـ GenAI ووجدت أن النية لاستخدام GenAI تأثرت بشكل إيجابي بتوقع الجهد، وقيمة السعر، والدافع الهيدوني. تأثرت توقعات الأداء، وتوقعات الجهد، وقيمة السعر، والدافع الهيدوني جميعها بشكل قوي بالتفاؤل والابتكار. تأثرت النية لاستخدام GenAI بشكل كبير بصفة الفضول. على الرغم من هذه الدراسات، لا توجد أبحاث كافية تستقصي كيف يمكن أن تتنبأ الجوانب الإيجابية والسلبية لأدوات الذكاء الاصطناعي التوليدي بشكل فعال بالمتغيرات الأساسية لنظرية السلوك المخطط (TPB)، مثل المواقف، والتحكم السلوكي المدرك، والمعايير الذاتية، والنوايا السلوكية. وبالتالي، تتناول المقالة الحالية هذه الفجوة البحثية من خلال توسيع نموذج TPB ليشمل سياق الذكاء الاصطناعي التوليدي.
تسعى الدراسة الحالية إلى كشف العوامل الرئيسية التي تدفع إلى اعتماد الذكاء الاصطناعي التوليدي في التعليم العالي. بشكل أكثر تحديدًا، تهدف هذه الورقة إلى (1) فحص تأثير نقاط القوة والفوائد والضعف والمخاطر المرتبطة باستخدام الذكاء الاصطناعي التوليدي على المواقف والمعايير الذاتية والتحكم السلوكي المدرك (PBC) للطلاب والمحاضرين؛ (2) التحقيق في تأثير المواقف والمعايير الذاتية وPBC للطلاب والمحاضرين على نيتهم في استخدام تطبيقات الذكاء الاصطناعي التوليدي؛ و(3) اختبار العلاقة بين نية الطلاب والمحاضرين واستخدامهم الفعلي لتطبيقات الذكاء الاصطناعي التوليدي في دراستهم و/أو أبحاثهم.
من خلال ذلك، تستكشف هذه الورقة كل من المزايا والعيوب التي قد تجلبها تقنيات الذكاء الاصطناعي التوليدي في الأبحاث والبيئات التعليمية. ونتيجة لذلك، يمكن أن يقوم المعنيون (مثل الإدارة العليا والمعلمين في مؤسسات التعليم العالي) بتصميم إرشادات وسياسات أفضل لاستغلال مزايا أدوات الذكاء الاصطناعي التوليدي مع تقليل أي سلبيات محتملة من خلال تطوير فهم شامل للتداعيات والصعوبات المحتملة لدمجها في الأبحاث والتعليم. بالإضافة إلى ذلك، تساعد هذه الدراسة في تحسين فهم السلوك البشري، خاصة فيما يتعلق بالتفاعل بين الإنسان والكمبيوتر، مما يسهل إنشاء تدخلات وتنظيمات تركز بشكل أكثر تحديدًا على تشجيع النتائج السلوكية المرغوبة في التعليم والبحث.
تنظم هذه الورقة على النحو التالي. يوفر القسم التالي مراجعة الأدبيات ويطور الفرضيات، بينما يوضح القسم الثالث المنهجية. يقدم القسم الرابع النتائج، ويناقش القسم الخامس الآثار النظرية والإدارية، ويحدد القسم الأخير قيود البحث واتجاهات البحث المستقبلية.

2. مراجعة الأدبيات وتطوير الفرضيات

2.1. نظرية السلوك المخطط

نظرية السلوك المخطط، التي قدمها أجن في أواخر الثمانينيات [20،21]، تقدم عدسة مفيدة لاستكشاف الديناميات الكامنة وراء السلوك البشري في سياق التعليم والتعلم، سواء بشكل عام أو في سياق استخدام التكنولوجيا كجزء من التعليم والبحث [22،23]. بشكل عام، تفترض نظرية السلوك المخطط أن السلوك يتشكل من خلال المواقف والمعايير الذاتية والرقابة السلوكية المدركة، والتي تؤثر بشكل جماعي على تشكيل النوايا السلوكية والإجراءات اللاحقة [24]. المركز في نظرية السلوك المخطط هو بناء المواقف، الذي يلتقط الأحكام التقييمية لسلوكيات مختلفة. في سياق التعليم والتعلم، فإن المواقف الإيجابية، المتجذرة في تصور إيجابي للنتائج المرتبطة بأنشطة التعليم والتعلم، وفقًا لنظرية السلوك المخطط، تدفع تطوير النوايا والانخراط التعليمي الناتج. تفحص المعايير الذاتية تأثير توقعات المجتمع والأقران على النية السلوكية والسلوك الفعلي [25]. إن الموافقة أو الرفض المدرك من قبل زملاء الدراسة، والزملاء، والإداريين، والمجتمع الأوسع حول التعليم العالي يعمل كعامل تحفيزي قوي، يوجه المعلمين والمتعلمين نحو التوافق مع المعايير التعليمية المدركة.
تقدم نظرية السلوك المخطط أيضًا مفهوم السيطرة السلوكية المدركة، والذي في سياق التعليم والتعلم يشير إلى المعتقدات الشاملة في القدرة على تنفيذ استراتيجيات التدريس بفعالية. من المتوقع أن تعزز السيطرة المدركة الأكبر كل من النية في الانخراط في ممارسات تدريس محددة والترجمة اللاحقة للنوايا إلى أفعال فعلية في الفصل الدراسي. تساهم هذه المكونات المترابطة مجتمعة في تشكيل النوايا السلوكية، التي تعتبر مقدمة حاسمة للسلوكيات الفعلية.
تجد مرونة نظرية السلوك المخطط (TPB) صدى في سياق التعليم والتعلم، كما يتضح من اعتمادها الواسع في الأبحاث التعليمية لدراسة، على سبيل المثال، فائدة الدورات التعليمية المفتوحة عبر الإنترنت [26] أو التعلم عبر الهواتف المحمولة [27]. لقد امتدت تطبيقاتها لتشمل جوانب متنوعة من البيداغوجيا، بدءًا من دمج تقنيات التعليم المبتكرة إلى اعتماد أساليب تعليمية تركز على الطالب مثل التعلم من الأقران [28]. في علم النفس التعليمي، تم استخدام نظرية السلوك المخطط للتحقيق في اعتماد المعلمين لممارسات التعليم المستندة إلى الأدلة، واستراتيجيات التقييم، وتقنيات إدارة الفصول الدراسية [22].
من خلال تحليل الروابط بين المواقف والمعايير الذاتية والإحساس بالتحكم، يقدم نموذج السلوك المخطط إطارًا لفهم اتخاذ القرارات التعليمية. إن تطبيقه يبرز إمكانيته ليس فقط في توضيح سلوكيات المعلمين والطلاب، ولكن أيضًا في توجيه الجهود لتصميم تدخلات مستهدفة، مثل استخدام الذكاء الاصطناعي التوليدي كجزء من التعليم والتعلم.

2.2. الذكاء الاصطناعي التوليدي في التعليم والتعلم

لقد ظهرت تقنية الذكاء الاصطناعي التوليدي كأداة تحويلية لها تداعيات متعددة الأوجه على التعليم. في مجال التدريس والتعلم، حظيت تقنية الذكاء الاصطناعي التوليدي باهتمام متزايد نظرًا لإمكاناتها في إعادة تشكيل الأساليب التربوية وتجارب التعلم في مجموعة متنوعة من المجالات، مثل العلوم الاجتماعية، والرياضيات، والهندسة. لمعالجة موضوع الذكاء الاصطناعي التوليدي في الفصول الدراسية، انضمت المؤسسات التعليمية الفردية وكذلك المنظمات التعليمية متعددة الجنسيات مثل اليونسكو إلى النقاش حول كيفية استخدام الذكاء الاصطناعي التوليدي بشكل أفضل في التدريس والتعلم. معظم المعلمين و
يبدو أن الباحثين يستنتجون أن ظهور الذكاء الاصطناعي التوليدي يمثل سيفًا ذو حدين، حيث أن من جهة، يوفر دمج الذكاء الاصطناعي التوليدي في التعليم مجموعة من الفوائد المحتملة لكل من المعلم والمتعلم، ولكن من جهة أخرى، فإنه يطرح أيضًا تحديات جديدة وإمكانية سوء الاستخدام.
فيما يتعلق بالفوائد، أشار دويفيدي وآخرون [4] إلى أن تجارب التعلم الشخصية، وهي حجر الزاوية في فلسفة التعليم المعاصرة، قد تتعزز بشكل كبير من خلال قدرات الذكاء الاصطناعي التوليدي. تتيح هذه التكنولوجيا إنشاء مواد تعليمية مخصصة للغاية، وتقييمات، وآليات تغذية راجعة، مما يلبي احتياجات الطلاب الفردية، ومتطلباتهم الخاصة، وتفضيلاتهم في التعلم. هذه الشخصية و قد يؤدي توفر معلم خاص إلى تعزيز المشاركة الأكبر وفهم أعمق لموضوع الدراسة [10]. ومع ذلك، قد يجد بعض الطلاب أن المعلمين الذكاء الاصطناعي أسهل في القبول من الآخرين، في حين يجب أن تكون البيانات التي تم تدريب الذكاء الاصطناعي عليها لتخصيص مواد التعلم تحت المراقبة المفتوحة أيضًا لتحسين ما تسميه وولمزلي [33] ‘الشفافية الوظيفية’.
بالنسبة للمعلم، تقدم الذكاء الاصطناعي التوليدي فرصًا لتبسيط وتحسين تطوير المواد التعليمية. غالبًا ما يستثمر المعلمون وقتًا وجهدًا كبيرين في إعداد موارد التعلم مثل الاختبارات والدروس وقوائم القراءة وأدلة الدراسة. يمكن أن يخفف الذكاء الاصطناعي التوليدي من هذا العبء من خلال أتمتة إنشاء مثل هذه المواد، مما يسمح للمعلمين بتخصيص المزيد من الوقت للتفاعل المباشر مع الطلاب. هذه الكفاءة في إنشاء المحتوى تحمل القدرة على تسريع العملية التعليمية مع الحفاظ على جودة التعليم، حيث قد يبدأ دور الباحث/المعلم في الانتقال من كونه منشئًا إلى كونه منسقًا للمعرفة، وهو ما يشبه الانتقال من كونه مؤلفًا إلى كونه محررًا في منشور علمي.
كمثال على فوائد التعلم الشخصي القائم على الذكاء الاصطناعي التوليدي، يمكن أن يستفيد تعلم اللغات، وهو مسعى معرفي معقد، من قدرة الذكاء الاصطناعي التوليدي على محاكاة تفاعلات اللغة في العالم الحقيقي (مثل الدردشة التفاعلية). غالبًا ما يواجه متعلمو اللغات صعوبة في العثور على فرص لممارسة اللغة بشكل غامر. يمكن أن يسد الذكاء الاصطناعي التوليدي، وخاصة النماذج اللغوية الكبيرة مثل GPT-4 أو LLAMA2، هذه الفجوة من خلال إنشاء سيناريوهات محادثة تشبه الواقع، مما يوفر للمتعلمين منصة ديناميكية قابلة للتكيف وشخصية لتحسين مهاراتهم اللغوية في سياقات حقيقية. قد تساهم هذه الممارسة اللغوية الغامرة – التي تمثل فعليًا شكلًا جديدًا من لعب الأدوار بين الإنسان والذكاء الاصطناعي – في تعزيز الكفاءة والثقة في التواصل.
بالإضافة إلى التجارب التعليمية المخصصة ومهارات تحرير النصوص (مثل: تدقيق القواعد، مراجعة النصوص، جعل الحجج أكثر إيجازًا، التفكير التكراري)، توفر أدوات الذكاء الاصطناعي الفرص لأنواع جديدة من التعبير الإبداعي ضمن السياقات التعليمية، حيث يمكن أن تُحفز المساعي الإبداعية للطلاب بواسطة أدوات مدعومة بالذكاء الاصطناعي تساعد في إنتاج محتوى جديد وخيالي. إن تعزيز القدرات الإبداعية لا يوسع آفاق التعليم فحسب، بل قد يساهم أيضًا في تطوير مهارات حل المشكلات والتفكير النقدي، بشرط أن يمتلك المعلمون المهارات اللازمة لتقديم تقنيات التعلم الجديدة بشكل استباقي وواثق في الفصول الدراسية.
على الرغم من العديد من الفوائد المحتملة، فإن دمج الذكاء الاصطناعي التوليدي في التعليم العالي ليس خاليًا من التحديات والاعتبارات الأخلاقية. على سبيل المثال، فإن التحيز في المحتوى الذي يتم إنشاؤه بواسطة الذكاء الاصطناعي يمثل مصدر قلق بالغ، حيث يمكن أن تستمر نماذج الذكاء الاصطناعي في تعزيز التحيزات الموجودة في مجموعة بيانات التدريب الخاصة بها، مما قد يعزز الصور النمطية ويهمش مجموعات سكانية معينة. يتطلب الحماية من التحيز وضمان العدالة في المحتوى التعليمي الذي يتم إنشاؤه بواسطة الذكاء الاصطناعي تدقيقًا واستراتيجيات تخفيف. كما يتطلب أيضًا أساليب قوية لاختيار واكتساب بيانات التدريب، حيث تعتمد فعالية نماذج اللغة الكبيرة بشكل كبير على توفر مجموعات بيانات عالية الجودة ومتوازنة لتدريب النموذج. نظرًا لأن مخرجات نظام الذكاء الاصطناعي التوليدي هي نتيجة مباشرة لتدريبه، هناك حاجة ملحة لوضع أطر لتدقيق عملية التدريب وبيانات التدريب لتطوير أنظمة ذكاء اصطناعي موثوقة.
مسألة الأصالة والتميز، جزء من نقاش أوسع حول قانون الملكية الفكرية بشأن المحتوى الذي يتم إنشاؤه بواسطة الذكاء الاصطناعي، تتطلب معالجة [37]. من يجب أن يمتلك حقوق المحتوى الذي يتم إنشاؤه بواسطة الذكاء الاصطناعي؟ من المسؤول عن المحتوى الذي يتخيله الذكاء الاصطناعي [40]؟ تحت أي ظروف يمكن استخدام البيانات التي ينشئها المستخدمون ولا يمكن استخدامها لتدريب نماذج GenAI الجديدة لاحقًا؟ هذه الأسئلة لم تتلق بعد إجابة نهائية. في سياق التعليم العالي، واحدة من المخاوف الرئيسية هي ظهور أنواع جديدة من الانتحال ونزاهة الأكاديمية [16]. تحقيق توازن بين مساعدة الذكاء الاصطناعي وتنمية مهارات التفكير النقدي المستقل لدى الطلاب، والدافع وجهد التعلم هو مسألة معقدة تتطلب نهجًا دقيقًا وابتكارًا تربويًا. علاوة على ذلك، فإن الفعالية التربوية لأدوات GenAI تتطلب أيضًا تقييمًا دقيقًا. بينما تحمل التكنولوجيا وعدًا في تعزيز تجارب التعلم، يجب فحص توافقها مع المبادئ التربوية المعمول بها بدقة [36]. يلعب المعلمون دورًا مهمًا في تحديد السياقات المناسبة لتنفيذ المحتوى الذي يتم إنشاؤه بواسطة الذكاء الاصطناعي وضمان أن يساهم هذا المحتوى بشكل فعال في الأهداف التعليمية [10]، كما هو محدد في نتائج التعلم (المستوى الدقيق) والمناهج الدراسية (المستوى الكلي). الميل لدى نماذج اللغة الكبيرة لتخيل [40] يتطلب أيضًا من المعلمين والمتعلمين تطوير مهارات تتعلق بالتفكير النقدي، ونقد وسائل الإعلام، والتحقق من الحقائق.
يمثل دمج GenAI في التعليم والتعلم تقدمًا كبيرًا مع فوائد محتملة تمتد من تجارب التعلم الشخصية إلى تسهيل إنشاء المحتوى وزيادة الإبداع. ومع ذلك، فإن هذه المزايا المحتملة مصحوبة بتحديات تتعلق بشكل خاص بحماية البيانات، والتحيز، والتمثيل، وقابلية تدقيق بيانات التدريب وعملية التدريب، وموثوقية المحتوى الذي يتم إنشاؤه بواسطة الذكاء الاصطناعي مقارنة بأشكال جديدة من الانتحال، والتوافق التربوي لتفاعلات المعلم-الذكاء الاصطناعي-الطالب، مثل التركيز بشكل أكبر على استجواب مخرجات النموذج بشكل نقدي. قد توفر تطبيقات GenAI معلومات خاطئة وتخترع حقائق، وقد يؤدي الاعتماد المفرط على تطبيقات الذكاء الاصطناعي لتنظيم المحتوى بدلاً من إنشائه إلى تقليل مهارات الطلاب. بينما تتنقل المؤسسات التعليمية وأصحاب المصلحة في دمج GenAI، فإن اتباع نهج حكيم ومستنير أخلاقيًا أمر ضروري.

2.3. تطوير الفرضيات

تشير الأدبيات الحالية إلى أن الذكاء الاصطناعي التوليدي يقدم فوائد تربوية متنوعة، بما في ذلك، على سبيل المثال لا الحصر، تجارب التعلم الشخصية وإنشاء المحتوى بكفاءة [4،29]. يُفترض أن هذه المزايا المدركة تشكل بشكل إيجابي مواقف المعلمين تجاه اعتماد تقنيات الذكاء الاصطناعي التوليدي، حيث تستند المواقف الإيجابية غالبًا إلى تصورات مواتية للنتائج [20،24]. في الوقت نفسه، من المحتمل أن تؤثر نقاط القوة المدركة للذكاء الاصطناعي التوليدي على المعايير الذاتية من خلال التوافق مع التوقعات المجتمعية والمؤسسية لأساليب التدريس المبتكرة، مما يعمل كعامل تحفيزي للاعتماد [25،26]. علاوة على ذلك، من المتوقع أن تعزز الكفاءات المكتسبة من خلال الذكاء الاصطناعي التوليدي السيطرة المدركة على استراتيجيات التدريس والتعلم، مما يتماشى مع تأكيد نظرية التخطيط السلوكي على دور السيطرة السلوكية المدركة في تشكيل النية والسلوك النهائي [22]. لذلك، يتم طرح الفرضيات التالية.
H1a. تؤثر القوة المدركة للذكاء الاصطناعي التوليدي بشكل إيجابي على الموقف تجاه استخدام الذكاء الاصطناعي التوليدي.
H1b. تؤثر القوة المدركة للذكاء الاصطناعي التوليدي بشكل إيجابي على المعايير الذاتية.
H1c. تُؤثِّر القوة المُدركة للذكاء الاصطناعي التوليدي بشكل إيجابي على السيطرة السلوكية المُدركة.
تحدد الأبحاث الحالية مجموعة من الفوائد المحتملة للذكاء الاصطناعي التوليدي، مثل تجارب التعلم الشخصية وتبسيط تطوير المواد التعليمية [9،29]. وفقًا لنظرية السلوك المخطط، يُفترض أن هذه الفوائد المدركة تؤثر بشكل إيجابي على المواقف تجاه الذكاء الاصطناعي التوليدي، حيث ترتبط المواقف عادةً بمدى تفضيل الفرد أو
تقييمات غير مواتية للنتائج السلوكية [24]. وبالمثل، إذا كانت فوائد GenAI تتماشى مع التوقعات التعليمية أو المجتمعية الأوسع، فمن المحتمل أن تؤثر بشكل إيجابي على المعايير الذاتية، مما يجعلها دافعًا تحفيزيًا لاعتماد التكنولوجيا [25]. وأخيرًا، يمكن أن تعزز الفوائد المتوقعة من GenAI، مثل أتمتة المهام التي تتطلب جهدًا كبيرًا، السيطرة السلوكية المدركة من خلال تعزيز المعتقدات في القدرة على تنفيذ هذه التكنولوجيا بنجاح في البيئات التعليمية [22]. بناءً على ذلك، نفترض أن.
H2a. الفوائد المدركة للذكاء الاصطناعي التوليدي لها تأثير إيجابي على الموقف تجاه استخدام الذكاء الاصطناعي التوليدي.
H2b. الفوائد المدركة للذكاء الاصطناعي التوليدي لها تأثير إيجابي على المعايير الذاتية.
H2c. الفوائد المدركة للذكاء الاصطناعي التوليدي لها تأثير إيجابي على السيطرة السلوكية المدركة.
بالإضافة إلى نقاط القوة والفوائد، تسلط الأدبيات الحالية الضوء على تحديات واعتبارات أخلاقية متنوعة تتعلق بنشر الذكاء الاصطناعي التوليدي في البيئات التعليمية، بما في ذلك قضايا حماية البيانات، والتحيز، والأصالة، والتوافق التربوي [16]. تشير نظرية التخطيط السلوكي إلى أن التصورات السلبية للنتائج السلوكية من المحتمل أن تؤثر سلبًا على المواقف تجاه سلوك معين [20]. وبالمثل، إذا كانت نقاط الضعف المدركة للذكاء الاصطناعي التوليدي لا تتماشى مع التوقعات الاجتماعية أو التعليمية، فمن المتوقع أن تؤثر هذه الفجوة سلبًا على المعايير الذاتية [25]. بالإضافة إلى ذلك، يمكن أن تؤدي نقاط الضعف المدركة مثل فقدان المهارات أو تعزيز التحيزات إلى تقويض الشعور بالتحكم السلوكي المدرك في تنفيذ التكنولوجيا بنجاح، كما تشير نظرية التخطيط السلوكي إلى أن التركيز على التحكم السلوكي المدرك هو عامل محدد للنوايا والأفعال [22]. وبالتالي، يتم طرح الفرضيات التالية.
H3a. نقاط الضعف المدركة للذكاء الاصطناعي التوليدي لها تأثير سلبي على الموقف تجاه استخدام الذكاء الاصطناعي التوليدي.
H3b. نقاط الضعف المدركة للذكاء الاصطناعي التوليدي لها تأثير سلبي على المعايير الذاتية.
H3c. نقاط الضعف المدركة للذكاء الاصطناعي التوليدي لها تأثير سلبي على التحكم السلوكي المدرك.
تم أيضًا إثارة العديد من المخاطر المتعلقة بتبني الذكاء الاصطناعي التوليدي في التعليم، مثل تعزيز التحيز، والمخاوف الأخلاقية حول الانتحال، وأسئلة حول النزاهة الأكاديمية. وفقًا لنظرية التخطيط السلوكي، تتأثر المواقف بتقييم النتائج السلوكية، ويمكن أن تؤدي المخاطر المرتبطة بالذكاء الاصطناعي التوليدي إلى توليد مواقف سلبية تجاه استخدامه [20]. وبالمثل، إذا كانت التوقعات الاجتماعية والتعليمية تتجنب المخاطر، فمن المعقول أن نفترض أن المخاطر المدركة ستؤثر سلبًا على المعايير الذاتية [25]. أخيرًا، تنص نظرية التخطيط السلوكي على أن التحكم السلوكي المدرك يتشكل من خلال المعتقدات في قدرة الفرد على أداء سلوك بنجاح؛ لذلك، قد تقوض المخاطر المدركة هذا الشعور بالتحكم، مما يقلل من احتمالية تنفيذ الذكاء الاصطناعي التوليدي في السياقات التعليمية [22]. وبالتالي، نفترض أن:
H4a. المخاطر المدركة للذكاء الاصطناعي التوليدي لها تأثير سلبي على الموقف تجاه استخدام الذكاء الاصطناعي التوليدي.
H4b. المخاطر المدركة للذكاء الاصطناعي التوليدي لها تأثير سلبي على المعايير الذاتية.
H4c. المخاطر المدركة للذكاء الاصطناعي التوليدي لها تأثير سلبي على التحكم السلوكي المدرك.
بشكل عام، تفترض نظرية التخطيط السلوكي أن الموقف، والمعايير الذاتية، والتحكم المدرك هي متغيرات سابقة تؤثر على النية السلوكية، والتي بدورها تؤدي إلى السلوك الفعلي [20]. تقترح H5 أن الموقف الإيجابي تجاه استخدام الذكاء الاصطناعي التوليدي سيؤثر إيجابيًا على النية لاستخدامه، وهو رابط تم إثباته تجريبيًا في سياقات متعددة ضمن أبحاث نظرية التخطيط السلوكي [23]. وبالمثل، تنص H6 على أن التأثيرات الاجتماعية وتأثيرات الأقران، الممثلة من خلال بناء المعايير الذاتية، تؤثر إيجابيًا على النية السلوكية، بما يتماشى مع
الأدبيات السابقة حول نظرية التخطيط السلوكي [25]. H7 توسع هذا من خلال افتراض أن التحكم السلوكي المدرك الأكبر، الذي يعكس المعتقدات في قدرة الفرد على تنفيذ سلوك، سيؤثر أيضًا بشكل إيجابي على النية لنشر الذكاء الاصطناعي التوليدي في البيئات التعليمية [22]. أخيرًا، H8 تختتم سلسلة السلوك من خلال اقتراح أن النية، كما تتأثر بالمتغيرات المذكورة أعلاه، ستؤثر إيجابيًا على الاستخدام الفعلي، وهو مبدأ أساسي في نظرية التخطيط السلوكي. وبالتالي، نفترض أن.
H5. الموقف تجاه استخدام الذكاء الاصطناعي التوليدي له تأثير إيجابي على النية لاستخدام الذكاء الاصطناعي التوليدي.
H6. المعايير الذاتية المتعلقة بالذكاء الاصطناعي التوليدي لها تأثير إيجابي على النية لاستخدام الذكاء الاصطناعي التوليدي.
H7. التحكم السلوكي المدرك بشأن الذكاء الاصطناعي التوليدي له تأثير إيجابي على النية لاستخدام الذكاء الاصطناعي التوليدي.
H8. النية لاستخدام الذكاء الاصطناعي التوليدي لها تأثير إيجابي على الاستخدام الفعلي للذكاء الاصطناعي التوليدي.
بشكل جماعي، تتماشى هذه الفرضيات مع المبادئ الأساسية لنظرية التخطيط السلوكي وتدعمها أيضًا التحديات والفرص المحددة التي تطرحها دمج الذكاء الاصطناعي التوليدي في السياقات التعليمية. الشكل 1 يقدم النموذج المفاهيمي لهذا البحث.

3. المنهجية

3.1. إجراءات أخذ العينات وجمع البيانات

تم جمع البيانات بين أبريل ويونيو 2023 باستخدام استبيان عبر الإنترنت مجهول تم تطويره على Google Forms. وقد مكن هذا الفترة من تأمين الحد الأدنى من حجم العينة. تم استخدام أخذ العينات غير الاحتمالية حيث لم يكن حجم السكان الإجمالي معروفًا لأن الدراسة الحالية استهدفت المحاضرين والطلاب حول العالم الذين استخدموا تطبيقات الذكاء الاصطناعي التوليدي. لتعزيز معدل الاستجابة وتجنب تحيز عدم الاستجابة، استخدم المؤلفون ثلاث تقنيات أخذ عينات. أولاً، استنادًا إلى تقنية أخذ العينات الملائمة، تم توزيع رابط الاستبيان على المشاركين عبر بريدهم الإلكتروني الشخصي. ثم، تم تطبيق أخذ العينات بالاختيار الذاتي من خلال نشر الرابط على منصات التواصل الاجتماعي. أخيرًا، استنادًا إلى تقنية أخذ العينات بالكرة الثلجية، شارك المؤلفون الرابط مع أفراد معروفين في دول مختلفة طالبين منهم مشاركة الرابط مع شبكاتهم. في المجموع، أكمل 543 مستجيبًا الاستبيان، تم استبعاد 240 منهم لأنهم أعلنوا أنهم لم يستخدموا أدوات الذكاء الاصطناعي التوليدي. من بين 303 استجابة المتبقية، تم إزالة 5 لأنها لم تكن كاملة. وبالتالي، تم استخدام 298 مستجيبًا (130 محاضرًا و168 طالبًا) من 47 دولة لمزيد من التحليل. ساهم استخدام عينة دولية في تنوع المستجيبين وإمكانية تعميم النتائج.
وفقًا للأعمال السابقة، من المهم ملاحظة أن الطريقة الأكثر شيوعًا لتحديد عدد المشاركين في توقعات النماذج الجزئية (PLS) هي أن تستند حجم العينة إلى عدد الانحدارات داخل إطار الدراسة (Barclay، Higgins، & Thompson، 1995). وفقًا لقاعدة العشر مرات، يجب أن يكون الحد الأدنى لحجم العينة أكثر من عشر مرات من أكبر المسارات الموجهة إلى متغير في النموذج بمستوى قوة 0.8 ومستوى دلالة 0.05 [42]. وفقًا لنموذج الدراسة الحالية، فإن حجم العينة المطلوب هو 40 مما يعني أن الاستجابات المجمعة لكلا المجموعتين كافية لتحليل PLS-SEM. يتم تقديم الخصائص الديموغرافية للعينة في الجدول 1.

3.2. تصميم الاستبيان والقياسات

تضمن الاستبيان ثلاثة أقسام. تضمن القسم الأول سؤال تصفية يميز المشاركين الذين استخدموا بالفعل تطبيقات الذكاء الاصطناعي التوليدي (على سبيل المثال، هل استخدمت تطبيق ذكاء اصطناعي توليدي في بحثك/دراستك؟ نعم/لا). لم يتمكن المشاركون الذين لم يستخدموا هذه الأدوات من ملء الأقسام المتبقية. القسم الثاني
الشكل 1. النموذج المفاهيمي.
تضمن بيانات سبع متغيرات في هذه الدراسة (تم قياسها على مقياس مستوى الاتفاق من 5 نقاط) وهي: نقاط قوة تطبيقات الذكاء الاصطناعي التوليدي، فوائد تطبيقات الذكاء الاصطناعي التوليدي، نقاط ضعف تطبيقات الذكاء الاصطناعي التوليدي، مخاطر تطبيقات الذكاء الاصطناعي التوليدي، الموقف، التحكم السلوكي المدرك، المعايير الذاتية، النية لاستخدام تطبيقات الذكاء الاصطناعي التوليدي، الاستخدام الفعلي لتطبيقات الذكاء الاصطناعي التوليدي. يقدم الملحق 1 البيانات و مصادرها [43-45]. كما تضمن القسم سؤالين عامين أيضًا: أسماء تطبيقات الذكاء الاصطناعي التوليدي المستخدمة وتكرار استخدام الذكاء الاصطناعي التوليدي حسب اتجاهات الاستخدام (تم قياسها من 1 – أبدًا إلى 5 – كثيرًا جدًا). تناول القسم الأخير البيانات الديموغرافية.

3.3. تحليل البيانات

SEM هي تقنية إحصائية تعتمد استراتيجية تأكيدية (أي اختبار الفرضيات) لفحص نظرية هيكلية حول حدث معين، كوسيلة للتحقيق في العلاقات المتصلة في نموذج معقد. تشمل SEM نهجين إحصائيين: SEM القائم على التباين (PLS-SEM) و SEM القائم على التغاير (CB-SEM). CB-SEM هي تقنية من SEM تتحقق عادةً من الفرضيات المقترحة أو تنفيها باستخدام برامج مثل AMOS و EQS و LISREL و MPlus. ومع ذلك، يفضل العديد من الباحثين في الأعمال استخدام PLS-SEM حيث قد لا يتم الحصول على توزيع طبيعي في الممارسة العملية، وتتطلب تقنية CB-SEM حجم عينة كبير.
تم استخدام PLS-SEM باستخدام Warp PLS V8 [46] في هذا البحث للأسباب التالية. أولاً، تعتبر PLS-SEM تقنية مفيدة لتقييم النماذج المعقدة. تُعتبر هذه الطريقة مناسبة لأنها ملائمة للنماذج المعقدة مثل نموذج الدراسة الحالية الذي يحتوي على تسعة متغيرات مع اتجاهات متعددة يمكن اختبارها في وقت واحد [47]، كما يُوصى بها لتقدير المتغيرات السلوكية [48]، وتُستخدم على نطاق واسع في بيئة التعليم العالي [49-51]. علاوة على ذلك، على عكس الطرق المعتمدة على التباين، يمكن لـ PLS-SEM معالجة البيانات ذات التوزيع غير الطبيعي، وهي ميزة حاسمة للدراسة الحالية. بالإضافة إلى ذلك، تتيح تفسير التباين بين مكونات الدراسة. تم تنفيذ خطوتين في PLS-SEM، وهما نموذج القياس والنموذج الهيكلي، حيث يتعامل الأول مع موثوقية وصلاحية المكونات بينما يقيس الثاني العلاقات بين المتغيرات [46].
استنادًا إلى WarpPLS، تم إجراء سلسلة من الاختبارات لتقييم ملاءمة النموذج لثلاثة نماذج معادلات هيكلية، والتي تم تقديمها في
الملحق 2. كانت جميع هذه التقييمات بوضوح ضمن النطاق المسموح به من القيم، مما يشير إلى أن نماذج المعادلات الهيكلية تطابقت مع بياناتها بشكل مرضٍ.

4. النتائج

4.1. نماذج القياس

في PLS-SEM، الخطوة الأولى هي تقييم النموذج الخارجي (المسمى أيضًا ‘نموذج القياس’) الذي يهدف إلى التحقق من موثوقية وصلاحية النموذج المقترح قبل فحص النموذج الداخلي (المسمى أيضًا ‘النموذج الهيكلي’) الذي يغطي اختبار الفرضيات المقترحة. للنجاح في هذه الخطوة، يجب تلبية شرطين للموثوقية وشرطين للصلاحية. تشمل الأولى موثوقية المؤشر وموثوقية الاتساق الداخلي. في هذا الصدد، كما هو موضح في الجدول 2، تم استبعاد جميع العناصر التي كانت تحميلها أقل من 0.7 كما اقترح Hair et al. [52]. وهذا يعني أنه تم تحقيق موثوقية المؤشر، مما يتيح التحقق من تدابير إضافية. ثم، تم تشغيل النموذج مرة أخرى مع جميع العناصر التي كانت تحميلها فوق هذا العتبة. بعد ذلك، تم التحقق من الاتساق الداخلي للتأكد من أن المستجيبين لديهم فهم متساوٍ للمقاييس المعطاة. يتم ذلك من خلال اختبار الموثوقية المركبة وألفا كرونباخ حيث كانت جميع القيم فوق 0.7، مما يؤكد أن العناصر المعتمدة تقيس البنى الخاصة بها (انظر الجدول 2).
أما بالنسبة للصلاحية، فقد تم فحص الصلاحية التقاربية والصلاحية التمييزية لضمان صلاحية النموذج. كانت قيمة AVE مرضية لجميع البنى لأنها كانت فوق القيمة الموصى بها 0.5 مما يؤكد الصلاحية التقاربية [52]. وهذا يشير إلى أن العناصر لكل متغير قادرة على توضيح أكثر من خمسين في المئة من التباين في المتغير الذي تتعلق به. الجداول عرض نتائج صلاحية التمييز من خلال اختبارين: فورنيل ولارك [53] وقياس التباين المتغاير – التباين الأحادي (HTMT) [54] للنماذج الثلاثة. تم تأكيد النهج الأول حيث أن الجذر التربيعي لـ AVE لكل بناء أعلى من الارتباطات مع البنى الأخرى. بالإضافة إلى ذلك، كانت جميع نتائج اختبار HTMT أقل من العتبة لـ مُميز بخط عريض. هذا يُظهر أن صلاحية التمييز ليست مهددة لأن كل متغير فريد عن الآخرين.
علاوة على ذلك، نظرًا لأن الإجابات تم الحصول عليها من نفس المصدر، تم فحص تباين الطريقة الشائعة (CMV) من خلال اختبارين.
الجدول 1
الخصائص السكانية للمشاركين.
التركيبة السكانية الفئات الطلاب محاضرون
ن % ن %
جنس ذكر ٨٠ ٤٧.٦ 69 53.1
أنثى ٨٨ 52.4 61 ٤٦.٩
عمر 18-30 ١١٦ 69.1 ٧ ٥.٤
31-40 ٣٢ 19 41 31.5
41-50 20 11.9 ٤٤ ٣٣.٨
51-60 0 0 ٢٤ 18.5
61+ 0 0 14 10.8
تعليمي الثانوية العامة أو أقل ٤٦ ٢٧.٤ ٥ 3.9
مستوى بكالوريوس 50 ٢٩.٧ 2 1.5
سيد 43 ٢٥.٦ ٢٢ 16.9
دكتوراه ٢٦ 15.5 98 75.4
الآخرون ٣ 1.8 ٣ 2.3
مجال الدراسة/ البحث العلوم الاجتماعية (مثل: الأعمال، الاقتصاد، السياحة والضيافة، علم النفس، القانون، إلخ) 134 79.8 ١١٠ 84.6
التكنولوجيا (مثل الهندسة، الروبوتات، علوم الحاسوب، الميكانيكا، إلخ) 15 8.9 12 9.2
الفنون والعلوم الإنسانية (مثل: العمارة، التاريخ، الأدب، الموسيقى، الفلسفة، إلخ) 12 7.1 ٤ 3.1
علوم الحياة والطب الحيوي (مثل: البيولوجيا، الطب، الزراعة، إلخ) ٥ ٣ 2 1.5
العلوم الفيزيائية (مثل علم الفلك، الكيمياء، الفيزياء، الرياضيات، إلخ) 2 1.2 2 1.5
دولة المشاركين بلغاريا 26 15.5 ٦ ٤.٦
إندونيسيا 20 11.9 ٨ 6.2
البرتغال ١٣ ٧.٧ 10 ٧.٧
الولايات المتحدة 10 6.0 ١٣ 10.0
فنلندا 15 8.9 ٥ 3.8
عمان 10 6.0 ٦ ٤.٦
بولندا 10 6.0 ٤ 3.1
المملكة المتحدة ٤ ٢.٤ 9 6.9
الهند ٥ 3.0 ٥ 3.8
باكستان ٤ ٢.٤ ٤ 3.1
إسبانيا 0 0.0 ٨ 6.2
ماليزيا ٥ 3.0 ٢ 1.5
مصر ٤ ٢.٤ ٣ 2.3
تركيا 2 1.2 ٥ 3.8
اليونان ٢ 1.2 ٤ 3.1
هولندا 2 1.2 ٤ 3.1
السويد 2 1.2 ٣ 2.3
تايوان ٣ 1.8 1 0.8
أستراليا 1 0.6 ٣ 2.3
آخرون 30 17.9 27 20.8
إجمالي 168 100.0 ١٣٠ 100.0
(تم تلبية نهج هارمان ذو العامل الواحد وقيم VIF للتعدد الخطي الكامل، مما يؤكد الاتساق الداخلي حيث كانت النسبة المئوية للتباين الكلي المفسر بواسطة عامل واحد أقل من 50% [55] وكانت قيم VIF أقل من 5، على التوالي [56]. وبالتالي، تم تلبية متطلبات نموذج القياس مما يسمح بالتحقق من الخطوة الثانية.)

4.2. النماذج الهيكلية

تم تقديم معاملات المسار لنماذج العينة الكلية، والمحاضرين، والطلاب في الأشكال 2-4، على التوالي. كانت قوة GenAI وفوائد GenAI لها تأثيرات إيجابية وملحوظة على العوامل الثلاثة لنموذج TPB (الموقف، المعايير الذاتية، والتحكم السلوكي المدرك) مع -القيم أقل من العتبة 0.05 في النماذج الثلاثة (انظر الجدول 6). لذلك، فإن H1a و H1b و H1c و H2a و H2b و H2c مدعومة في النماذج الثلاثة.
أظهرت النتائج أن تأثير نقاط ضعف GenAI على
ظل الموقف غير ذي دلالة في النماذج الثلاثة جميعها كما كانت قيم – أعلى من العتبة 0.05 كما هو موضح في الأشكال 2-4: العينة الكلية ( ; المحاضرون ) ، والطلاب ( 0.26). وبالمثل، ظل تأثير نقاط ضعف GenAI على المعايير الذاتية غير ذي دلالة في النماذج الثلاثة: العينة الكلية ( ; المحاضرون ) ، والطلاب ( 0.44). بالإضافة إلى ذلك، لم يتم دعم تأثيره السلبي على السيطرة السلوكية المدركة في النماذج الثلاثة: العينة الكلية ( المحاضرون )، الطلاب ( ) (الأشكال 2 و 4). لذلك، تم رفض H3a و H3b و H3c.
بالإضافة إلى ذلك، أشارت النتائج إلى أن المخاطر المدركة للذكاء الاصطناعي التوليدي كان لها تأثير كبير فقط على المواقف في النموذج العام. ) ونموذج المحاضرين ( )؛ ومع ذلك، لم يكن هناك علاقة ذات دلالة في نموذج الطلاب ( ; ص ). وبالتالي تم قبول H4a لكل من عينة المحاضرين والعينة الكلية. بالإضافة إلى ذلك، لم تدعم النتائج العلاقة السلبية والمعنوية بين المخاطر المدركة للذكاء الاصطناعي التوليدي والمعايير الذاتية في النماذج الثلاثة: العينة الكلية ( المحاضرون ) ، والطلاب ( ). وبالتالي، تم رفض H4b. كان تأثير المخاطر المدركة للذكاء الاصطناعي التوليدي على السيطرة السلوكية المدركة كبيرًا في النموذج العام ( ; 0.04) ولكن غير مهم للمحاضرين ( ) والطلاب ( لذلك، تم قبول H4c للعينة الكاملة فقط.
أشارت النتائج التجريبية أيضًا إلى أن العلاقات بين المتغيرات الثلاثة الأساسية لنموذج نظرية التخطيط السلوكي (أي، الاتجاه، المعايير الذاتية، والتحكم السلوكي المدرك) ونية استخدام أدوات الذكاء الاصطناعي كانت ذات دلالة وإيجابية في النماذج الثلاثة. وبالتالي، تم قبول H5 وH6 وH7 جميعًا. أخيرًا، كانت هناك علاقة ذات دلالة وإيجابية بين النية للاستخدام والاستخدام الفعلي للذكاء الاصطناعي في النماذج الثلاثة. وبالتالي، تم دعم H8 (انظر الجدول 6).

5. المناقشة والاستنتاجات

5.1. المناقشة والآثار النظرية

مدعومًا بنموذج نظرية التخطيط السلوكي، بحثت هذه الدراسة العلاقة بين القوى المدركة، المزايا، الضعف، والمخاطر لأدوات الذكاء الاصطناعي التوليدي والمتغيرات الأساسية لنموذج نظرية التخطيط السلوكي، وهي الموقف، المعايير الذاتية، والتحكم السلوكي المدرك. كما استكشفت هذه الورقة كيف يمكن أن تؤثر المتغيرات الأساسية لنموذج نظرية التخطيط السلوكي على النية نحو اعتماد أدوات الذكاء الاصطناعي التوليدي، وكيف يمكن أن تؤثر هذه الأخيرة على الاستخدام الفعلي لأدوات الذكاء الاصطناعي التوليدي من قبل المحاضرين والطلاب في مؤسسات التعليم العالي. بشكل عام، تأثرت مواقف المحاضرين والطلاب، والمعايير الذاتية، والتحكم السلوكي المدرك بشكل كبير وإيجابي بالقوى المدركة والفوائد لأدوات الذكاء الاصطناعي التوليدي؛ ومع ذلك، تختلف تصورات المحاضرين والطلاب حول نقاط الضعف والمخاطر لهذه الأدوات.
تشير النتائج التجريبية إلى أن نقاط القوة والفوائد لتطبيقات الذكاء الاصطناعي التوليدي كان لها تأثير إيجابي وملحوظ على جميع مكونات نظرية السلوك المخطط الثلاثة (أي، الموقف، المعايير الذاتية، والتحكم السلوكي المدرك) في النماذج الثلاثة (العينة الكلية، المحاضرون، والطلاب). تشير هذه النتائج إلى أنه كلما كانت نقاط القوة والفوائد المتاحة لتقنيات الذكاء الاصطناعي التوليدي المستخدمة في مؤسسات التعليم العالي أكثر قابلية للتحقيق، زادت التأثيرات الإيجابية التي تحدثها على مواقف المحاضرين والطلاب، والمعايير الذاتية، والتحكم السلوكي المدرك. تؤكد هذه النتائج على أهمية عرض مزايا وخصائص تطبيقات الذكاء الاصطناعي التوليدي لتعزيز المواقف الإيجابية، والمعايير الذاتية، وإدراك المستخدمين لمدى قابليتها للاستخدام – وكل ذلك يمكن أن يؤدي في النهاية إلى قبول واستخدام فعال لهذه الأدوات في مؤسسات التعليم العالي. تدعم هذه النتائج ما توصلت إليه الدراسات السابقة التي تشير إلى أن نقاط القوة والفوائد المدركة للتقنيات يمكن أن تشكل بشكل إيجابي مواقف المعلمين (على سبيل المثال، المرجع [24])، والمعايير الذاتية (على سبيل المثال، المرجع [26])، والتحكم المدرك (على سبيل المثال، المرجع [22]) تجاه اعتماد مثل هذه الأدوات.
لم تدعم النتائج التجريبية بعض الفرضيات المتعلقة بالصلات بين نقاط ضعف الذكاء الاصطناعي التوليدي و
الجدول 2
الإحصائيات الوصفية ونتائج التقييم لنماذج القياس.
إجمالي العينة ( ) محاضرون ) الطلاب )
إنشاء/عناصر تحميل المؤشر الموثوقية المركبة AVE تحميل المؤشر الموثوقية المركبة AVE تحميل المؤشر الموثوقية المركبة AVE
قوة الذكاء الاصطناعي العام 0.876 0.640 0.811 0.799 0.883 0.655
STRN1 0.768 0.794 0.752
STRN2 0.850 0.804 0.840
STRN3 0.826 0.794 0.879
STRN4 0.752 0.804 0.760
فوائد الذكاء الاصطناعي التوليدي 0.938 0.791 0.915 0.893 0.938 0.790
BNFT1 0.892 0.916 0.878
BNFT2 0.863 0.839 0.879
بي إن إف تي 3 0.889 0.902 0.881
بي إن إف تي 4 0.913 0.915 0.916
نقاط ضعف الذكاء الاصطناعي العام 0.873 0.580 0.866 0.776 0.854 0.593
ضعيف1 0.816 0.795 0.782
ضعيف2 0.822 0.877 0.766
WEAK3 0.719 0.780 0.770
WEAK4 0.740 0.708 0.762
ويك5 0.702 0.745 غير متوفر
WEAK6 غير متوفر 0.739 غير متوفر
مخاطر الذكاء الاصطناعي التوليدي 0.918 0.848 0.858 0.936 0.863 0.678
RSK1 0.921 0.936 0.863
RSK2 0.921 0.936 0.857
RSK3 غير متوفر غير متوفر 0.745
الموقف 0.932 0.697 0.931 0.865 0.933 0.665
ATTD1 0.787 0.747 0.830
ATTD2 0.843 0.856 0.818
ATTD3 0.845 0.882 0.785
ATTD4 0.848 0.924 0.804
ATTD5 0.859 0.872 0.872
ATTD6 0.825 0.896 0.769
ATTD7 غير متوفر غير متوفر 0.827
المعايير الذاتية 0.930 0.816 0.921 0.929 0.909 0.770
الموضوع 1 0.906 0.930 0.882
SUBJ2 0.912 0.928 0.897
SUBJ3 0.891 0.930 0.851
التحكم السلوكي المدرك 0.893 0.736 0.827 0.862 0.892 0.733
PRCV1 0.827 0.817 0.836
PRCV2 0.867 0.884 0.854
PRCV3 0.879 0.884 0.877
نية استخدام الذكاء الاصطناعي التوليدي 0.898 0.747 0.857 0.882 0.887 0.724
INTN1 0.856 0.876 0.843
INTN2 0.872 0.884 0.862
INTN3 0.864 0.886 0.848
الاستخدام الفعلي للذكاء الاصطناعي التوليدي 0.922 0.855 0.858 0.936 0.909 0.833
ACTU1 0.924 0.936 0.913
ACTU2 0.924 0.936 0.913
ثلاثة متغيرات من نظرية السلوك المخطط. على الرغم من أن الدراسات السابقة قد أكدت على نقاط ضعف GenAI وفرص إساءة الاستخدام (مثل المرجع [15،16])، إلا أن هذه الدراسة وجدت أنها لم تكن حاسمة في تحديد مواقف المحاضرين والطلاب، والمعايير الاجتماعية، أو تصوراتهم حول السيطرة السلوكية، بل كانت نقاط القوة والفوائد لـ GenAI هي الحاسمة. بالإضافة إلى ذلك، أنتج التحليل نتائج مختلطة بشأن تأثير المخاطر المدركة لـ GenAI على الموقف، والمعايير الذاتية، والسيطرة السلوكية المدركة (فقط 3 من 9 فرضيات تتعلق بالمخاطر المدركة تم دعمها في النماذج الثلاثة). وهذا يعني أن المحاضرين والطلاب يختلفون في المواقف، وتصورات السيطرة السلوكية، والالتزام بالمعايير الاجتماعية عندما يتعلق الأمر بكيفية فهمهم والتفاعل مع المخاطر المحيطة بتطبيقات GenAI (مثل ChatGPT). في هذا السياق، توفر النتائج دعمًا جزئيًا فقط للدراسات السابقة (مثل المرجع [4،9،16،37]) التي سلطت الضوء على المخاطر المدركة المتعلقة بتبني تطبيقات GenAI، بما في ذلك ChatGPT. على الرغم من أن الأدبيات الحالية (مثل المرجع [16]) قد أوضحت العديد من الصعوبات والمخاوف الأخلاقية المحيطة بتطبيق GenAI في بيئات التعلم، مثل أمان البيانات، والتحيز، والأصالة، والتماسك التربوي، أظهرت هذه الدراسة أن هذه المخاطر ليست مهمة جدًا بالنسبة للمحاضرين والطلاب. في الواقع، فإن النقاط القوية والفوائد المدركة لاستخدام GenAI في التعليم والبحث هي محركات أكثر أهمية لتبني GenAI في البيئة التعليمية من نقاط ضعفها ومخاطرها. إن حداثة هذه الأدوات، والإمكانات
المزايا التي تقدمها للطلاب والمحاضرين من حيث توفير الوقت وزيادة الإنتاجية، والتحسينات الكبيرة في جودة مخرجات الذكاء الاصطناعي التوليدي مع مرور الوقت قد تكون الأسباب التي تجعل نقاط القوة والفوائد للذكاء الاصطناعي التوليدي أكثر أهمية بالنسبة للمستجيبين من نقاط الضعف والمخاطر المرتبطة بهذه الأدوات، ولكن يجب أن تقدم الأبحاث المستقبلية إجابة حاسمة على هذا السؤال.
أظهرت النتائج أيضًا أن المتغيرات الثلاثة الأساسية لنموذج السلوك المخطط (أي، الموقف، المعايير الذاتية، والتحكم السلوكي المدرك) كانت لها علاقة إيجابية وملحوظة مع نية استخدام أدوات الذكاء الاصطناعي التوليدي من قبل كل من المحاضرين والطلاب. لتشجيع استخدام ودمج أدوات الذكاء الاصطناعي التوليدي في التعليم العالي، من الضروري تنمية مواقف إيجابية، ومعايير ذاتية، وتحكم سلوكي مدرك تجاه هذه الأدوات بين كل من المحاضرين والطلاب. قد تزداد نوايا كل من المحاضرين والطلاب لاستخدام هذه الأدوات من خلال تسليط الضوء على المزايا والفوائد والفعالية لهذه الأدوات. سيساهم ذلك في تحسين مواقف الطلاب والمحاضرين، والمعايير الذاتية، والتحكم المدرك. تتماشى هذه النتائج مع مبادئ نموذج السلوك المخطط، الذي يؤكد أن نية الشخص في الانخراط في سلوك معين – في هذه الحالة، استخدام أدوات الذكاء الاصطناعي التوليدي – تتأثر بشكل كبير بهذه المتغيرات الثلاثة. كما أظهرت النتائج وجود علاقة قوية بين نية استخدام تقنيات الذكاء الاصطناعي التوليدي وتطبيقها الفعلي. تتماشى هذه النتيجة مع الفهم العام في السلوكيات.
الجدول 3
صلاحية التمييز (العينة الكلية).
بناء فورنيل ولاركَر [53]
ستران بي إن إف تي ضعيف RSK ATTD الموضوع PRCV تهيئة ACTU
ستران (0.800)*
بي إن إف تي 0.657 (0.889)
ضعيف -0.079 -0.017 (0.762)
RSK -0.058 -0.062 0.334 (0.921)
ATTD 0.554 0.575 -0.056 -0.149 (0.835)
الموضوع 0.335 0.458 0.007 -0.039 0.420 (0.903)
PRCV 0.567 0.591 0.084 -0.032 0.503 0.379 (0.858)
INTN 0.586 0.668 -0.009 -0.095 0.582 0.527 0.709 (0.864)
ACTU 0.398 0.491 0.012 -0.111 0.472 0.506 0.540 0.684 (0.924)
نسب HTMT STRN بي إن إف تي ضعيف RSK ATTD الموضوع PRCV تهيئة ACTU
ستران
بي إن إف تي 0.766**
ضعيف 0.168 0.102
RSK 0.075 0.078 0.408
ATTD 0.645 0.631 0.144 0.172
الموضوع 0.394 0.509 0.066 0.053 0.468
PRCV 0.698 0.680 0.166 0.088 0.580 0.442
INTN 0.717 0.767 0.139 0.115 0.670 0.615 0.858
ACTU 0.487 0.564 0.056 0.135 0.543 0.590 0.651 0.823
*الأرقام بين الأقواس تعكس الجذر التربيعي للقيم المتوسطة (AVEs)، بينما تشير الأرقام الأخرى إلى الارتباطات بين العوامل.
** القيم الجريئة لنسبة HTMT التي تقل عن 0.90 تشير إلى أن: تلك المتغيرات متميزة عن المتغيرات الأخرى مما يؤكد تفردها.
الجدول 4
صلاحية التمييز (المحاضرون).
بناء فورنيل ولاركَر [53]
ستران بي إن إف تي ضعيف RSK ATTD الموضوع PRCV تهيئة ACTU
ستران (0.799)
بي إن إف تي 0.549 (0.893)
ضعيف -0.112 -0.009 (0.776)
RSK -0.167 -0.187 0.282 (0.936)
ATTD 0.596 0.536 -0.062 -0.360 (0.865)
الموضوع 0.٤٥١ 0.535 0.051 -0.154 0.416 (0.862)
PRCV 0.516 0.657 -0.033 -0.244 0.602 0.638 (0.882)
INTN 0.460 0.586 -0.013 -0.251 0.526 0.589 0.806 (0.936)
ACTU 0.297 0.452 0.008 -0.174 0.426 0.303 0.488 0.527 (0.814)
نسب HTMT بي إن إف تي ضعيف RSK ATTD الموضوع PRCV تهيئة ACTU
ستران
بي إن إف تي 0.638
ضعيف 0.159 0.088
RSK 0.200 0.211 0.330
ATTD 0.686 0.582 0.127 0.405
الموضوع 0.549 0.610 0.193 0.180 0.472
PRCV 0.618 0.741 0.093 0.285 0.676 0.754
INTN 0.551 0.661 0.094 0.293 0.590 0.695 0.940
ACTU 0.360 0.519 0.098 0.336 0.495 0.353 0.573 0.618
*الأرقام بين الأقواس تعكس الجذر التربيعي للقيم المتوسطة (AVEs)، بينما تشير الأرقام الأخرى إلى الارتباطات بين العوامل.
** القيم الجريئة لنسبة HTMT التي تقل عن 0.90 تشير إلى أن: تلك المتغيرات متميزة عن المتغيرات الأخرى مما يؤكد تفردها.
علم النفس يشير إلى أن النوايا غالبًا ما تكون محددًا موثوقًا للسلوك اللاحق. إنها نظرة قيمة لفهم العوامل التي تؤثر على اعتماد أدوات الذكاء الاصطناعي التوليدي في الممارسة من قبل كل من المحاضرين والطلاب (على سبيل المثال، المرجع [57،58]).
من الناحية النظرية، يكمل هذا العمل الأبحاث السابقة التي أكدت على القضايا الرئيسية المحيطة بتطبيق أدوات الذكاء الاصطناعي التوليدي في البحث والتعليم [9،37]. ركزت الدراسات السابقة في هذا المجال على المتغيرات الرئيسية التي تؤثر على استخدام هذه الأدوات في البحث والتعليم ضمن بيئات مختلفة (على سبيل المثال، المرجع [57-59]). ومع ذلك، فإن الدراسة الحالية هي واحدة من أولى المحاولات لدمج عدة عناصر حاسمة، مثل نقاط القوة والفوائد والضعف والمخاطر المرتبطة باستخدام أدوات الذكاء الاصطناعي التوليدي ضمن نموذج هيكلي لقياس العوامل الأكثر أهمية التي تؤثر على الاستخدام الفعلي لهذه الأدوات استنادًا إلى عينة دولية من المحاضرين/الباحثين والطلاب. علاوة على ذلك، تعزز هذه الدراسة النظرية من خلال توسيع نموذج نظرية التخطيط السلوكي لتوضيح المحركات الرئيسية لنشر تطبيقات الذكاء الاصطناعي التوليدي مثل
تشات جي بي تي في مؤسسات التعليم العالي. بالإضافة إلى ذلك، هذه واحدة من الدراسات الأولى التي تقارن آراء المحاضرين والطلاب حول العناصر الحاسمة التي تؤثر على استخدام أدوات الذكاء الاصطناعي التوليدي.

5.2. الآثار الإدارية والسياسية

تقدم الدراسة مجموعة من الآثار العملية للجهات المعنية في مؤسسات التعليم العالي (مثل الطلاب والمحاضرين وإداريي مؤسسات التعليم العالي). تكشف النتائج التجريبية أن تحسين وجهات نظر المحاضرين والطلاب حول مزايا وفوائد تنفيذ أدوات الذكاء الاصطناعي التوليدي يمكن أن يرتبط بموقف أكثر إيجابية، ومعايير ذاتية، وإدراك للسيطرة السلوكية تجاه استخدام مثل هذه الأدوات في البحث والتعليم في مؤسسات التعليم العالي. يمكن تحقيق ذلك من خلال تطوير ونشر إجراءات وآليات فعالة لنشر المعرفة والوعي حول استخدام أدوات الذكاء الاصطناعي التوليدي في التعليم والبحث. بالإضافة إلى ذلك، تحتاج مؤسسات التعليم العالي إلى تنظيم ورش عمل وبناء الوعي.
الجدول 5
صلاحية التمييز (الطلاب).
بناء فورنيل ولاركَر [53]
ستران بي إن إف تي ضعيف RSK ATTD الموضوع PRCV تهيئة ACTU
ستران (0.712)
بي إن إف تي 0.688 (0.889)
ضعيف -0.077 0.002 (0.692)
RSK 0.054 0.077 0.489 (0.663)
ATTD 0.537 0.601 0.004 0.084 (0.816)
الموضوع 0.417 0.457 0.038 0.217 0.388 (0.877)
PRCV 0.605 0.679 0.021 0.111 0.562 0.551 (0.851)
INTN 0.349 0.426 0.044 0.084 0.414 0.457 0.576 (0.913)
ACTU 0.620 0.629 0.115 0.148 0.582 0.448 0.766 0.499 (0.856)
نسب HTMT STRN بي إن إف تي ضعيف RSK ATTD الموضوع PRCV تهيئة ACTU
ستران
بي إن إف تي 0.745
ضعيف 0.183 0.106
RSK 0.066 0.120 0.569
ATTD 0.560 0.657 0.148 0.096
الموضوع 0.493 0.520 0.158 0.180 0.441
PRCV 0.693 0.791 0.125 0.097 0.652 0.666
INTN 0.341 0.500 0.084 0.113 0.484 0.554 0.715
ACTU 0.691 0.728 0.139 0.137 0.672 0.538 0.941 0.615
*الأرقام بين الأقواس تعكس الجذر التربيعي للقيم المتوسطة (AVEs)، بينما تشير الأرقام الأخرى إلى الارتباطات بين العوامل.
** القيم الجريئة لنسبة HTMT التي تقل عن 0.90 تشير إلى أن: تلك المتغيرات متميزة عن المتغيرات الأخرى مما يؤكد تفردها.
الشكل 2. النموذج الهيكلي (العينة الكلية).
الشكل 3. النموذج الهيكلي (المحاضرون).
الشكل 4. النموذج الهيكلي (الطلاب).
الجدول 6
ملخص نتائج الفرضيات.
فرضية نتيجة
العينة الكلية محاضرون الطلاب
H1a تؤثر القوة المدركة للذكاء الاصطناعي التوليدي بشكل إيجابي على الموقف تجاه استخدام الذكاء الاصطناعي التوليدي. مقبول مقبول مقبول
H1b تؤثر القوة المدركة للذكاء الاصطناعي التوليدي بشكل إيجابي على المعايير الذاتية. مقبول مقبول مقبول
H1c تؤثر القوة المدركة للذكاء الاصطناعي التوليدي بشكل إيجابي على السيطرة السلوكية المدركة. مقبول مقبول مقبول
H2a تؤثر الفوائد المدركة للذكاء الاصطناعي التوليدي بشكل إيجابي على الموقف تجاه استخدام الذكاء الاصطناعي التوليدي. مقبول مقبول مقبول
H2b تؤثر الفوائد المدركة للذكاء الاصطناعي التوليدي بشكل إيجابي على المعايير الذاتية. مقبول مقبول مقبول
H2c تؤثر الفوائد المدركة للذكاء الاصطناعي التوليدي بشكل إيجابي على السيطرة السلوكية المدركة. مقبول مقبول مقبول
H3a تؤثر نقاط الضعف المتصورة في الذكاء الاصطناعي التوليدي سلبًا على الموقف تجاه استخدام الذكاء الاصطناعي التوليدي. مرفوض مرفوض مرفوض
H3b تؤثر نقاط الضعف المتصورة للذكاء الاصطناعي التوليدي سلبًا على المعايير الذاتية. مرفوض مرفوض مرفوض
H3c تؤثر نقاط الضعف المتصورة في الذكاء الاصطناعي التوليدي سلبًا على السيطرة السلوكية المتصورة. مرفوض مرفوض مرفوض
H4a تؤثر المخاطر المدركة للذكاء الاصطناعي التوليدي سلبًا على الموقف تجاه استخدام الذكاء الاصطناعي التوليدي. مقبول مقبول مرفوض
H4b تؤثر المخاطر المدركة للذكاء الاصطناعي التوليدي سلبًا على المعايير الذاتية. مرفوض مرفوض مرفوض
H4c تؤثر المخاطر المدركة للذكاء الاصطناعي التوليدي سلبًا على السيطرة السلوكية المدركة. مقبول مرفوض مرفوض
H5 الموقف تجاه استخدام الذكاء الاصطناعي التوليدي له تأثير إيجابي على النية لاستخدام الذكاء الاصطناعي التوليدي مقبول مقبول مقبول
H6 المعايير الذاتية المتعلقة بالذكاء الاصطناعي التوليدي لها تأثير إيجابي على النية لاستخدام الذكاء الاصطناعي التوليدي مقبول مقبول مقبول
H7 التحكم المدرك فيما يتعلق بالذكاء الاصطناعي التوليدي له تأثير إيجابي على نية استخدام الذكاء الاصطناعي التوليدي مقبول مقبول مقبول
H8 نية استخدام الذكاء الاصطناعي التوليدي لها تأثير إيجابي على الاستخدام الفعلي للذكاء الاصطناعي التوليدي مقبول مقبول مقبول
مبادرات وجلسات تدريبية تُظهر نقاط القوة والفوائد من اعتماد تطبيقات الذكاء الاصطناعي التوليدي، بما في ذلك ChatGPT، لأغراض التعليم والبحث.
تسلط النتائج الضوء على دور الموقف والمعايير الذاتية والرقابة السلوكية المدركة في تشكيل اعتماد أدوات الذكاء الاصطناعي التوليدي في التعليم والبحث. تزيد المواقف الإيجابية تجاه أدوات الذكاء الاصطناعي التوليدي بين الأكاديميين والطلاب من احتمال استخدامهم لهذه الأدوات، مما يبرز أهمية تعزيز المواقف الإيجابية من خلال البرامج التعليمية والتأكيد على مزايا هذه الأدوات في كل من التعليم والبحث. علاوة على ذلك، يمكن أن تؤثر المعايير الذاتية التي تشكلها آراء المجتمعات الأكاديمية والأقران على مدى قبول أدوات الذكاء الاصطناعي التوليدي اجتماعيًا. يجب على مؤسسات التعليم العالي تعزيز جو تعاوني يشجع على تبادل المعرفة ويبرز الفوائد المشتركة لاستخدام أدوات الذكاء الاصطناعي التوليدي في بيئات البحث والتعليم. بالإضافة إلى ذلك، تشمل الرقابة السلوكية المدركة جوانب مثل سهولة الاستخدام المدركة والخبرة التقنية. لتحسين ثقة المستخدمين في استخدام أدوات الذكاء الاصطناعي التوليدي بكفاءة، يجب على مؤسسات التعليم العالي تطوير معايير أخلاقية للاستخدام المسؤول للذكاء الاصطناعي التوليدي وتقديم برامج تدريبية ذات صلة. تحتاج مؤسسات التعليم العالي إلى إنشاء ثقافة من القبول والكفاءة بين المحاضرين والطلاب من خلال إنشاء بيئة ملائمة للتكامل الفعال والكفء والأخلاقي لأدوات الذكاء الاصطناعي التوليدي في التعليم والبحث. تحتاج مؤسسات التعليم العالي إلى تطوير مبادرات لتشجيع المحاضرين والطلاب على استخدام الذكاء الاصطناعي التوليدي بشكل مسؤول. يمكن لصانعي السياسات المساعدة من خلال تطوير تنظيمات قانونية لـ
إدارة تطبيق الذكاء الاصطناعي العام في البحث والتعليم للتخفيف من آثاره السلبية من خلال إشراك مؤسسات التعليم العالي في العملية. علاوة على ذلك، لدعم الشمولية والتنوع في البحث والتعليم، تحتاج السلطات العامة ومؤسسات التعليم العالي إلى ضمان توفر تقنيات الذكاء الاصطناعي العام لجميع المحاضرين والطلاب لضمان قدرتهم التنافسية وقابلية توظيفهم، على سبيل المثال، من خلال حسابات مؤسسية لتطبيقات الذكاء الاصطناعي العام.

5.3. القيود واتجاهات البحث المستقبلية

على الرغم من أن الدراسة الحالية تقدم عدة مساهمات نظرية وعملية، إلا أن بعض القيود توفر اتجاهات قيمة للبحوث المستقبلية. أولاً، اعتمدت الدراسة الحالية على نهج كمي باستخدام استبيان عبر الإنترنت لجمع البيانات الأولية من المستجيبين المستهدفين (أي الطلاب والمحاضرين) من عدة دول. يمكن أن تعتمد الأبحاث المستقبلية نهجًا نوعيًا من خلال استخدام المقابلات أو نهج مختلط. ثانيًا، يعتمد نموذج الدراسة الحالية على نظرية السلوك المخطط (TPB) ويقوم بتوسيعها من خلال تضمين مجموعة من المتغيرات (أي نقاط قوة الذكاء الاصطناعي التوليدي، فوائد الذكاء الاصطناعي التوليدي، نقاط ضعف الذكاء الاصطناعي التوليدي، ومخاطر الذكاء الاصطناعي التوليدي). يمكن أن تطور الدراسات المستقبلية نموذجًا أكثر شمولاً من خلال دمج نظريات أخرى مثل نظرية انتشار الابتكار، ونموذج قبول التكنولوجيا، أو النظرية الموحدة لقبول واستخدام التكنولوجيا. يمكن أيضًا دمج متغيرات إضافية، مثل الضغط، والثقة، والكفاءة الذاتية، في النموذج الحالي.
ثالثًا، ركزت هذه الدراسة على التعليم العالي. يمكن أن تركز الأبحاث المستقبلية على استخدام الذكاء الاصطناعي التوليدي في التعليم الثانوي للحصول على رؤى حول المتغيرات التي تؤثر على تصورات وسلوكيات طلاب المدارس الثانوية والمعلمين تجاه تطبيقات أدوات الذكاء الاصطناعي التوليدي. بالإضافة إلى ذلك، نظرًا لأن استخدام أدوات الذكاء الاصطناعي التوليدي هو المتغير الناتج في الدراسة، يمكن أن تبحث الأبحاث المستقبلية في آثار هذا الاستخدام في مجموعة متنوعة من المجالات، بما في ذلك مستويات التفكير الإبداعي، والإنجاز الأكاديمي، والإنتاجية العلمية، وأخلاقيات البحث. علاوة على ذلك، قد تتناول الأبحاث المستقبلية تفاصيل استخدام الذكاء الاصطناعي التوليدي في دول مختلفة، مما يسمح بفهم أكثر شمولاً للروابط المحتملة والأنماط المميزة عبر السياقات الثقافية والاجتماعية والاقتصادية المتنوعة.

إعلان الذكاء الاصطناعي التوليدي في الكتابة العلمية

خلال إعداد هذا العمل، استخدم المؤلفون ChatGPT وGrammarly لتدقيق النص وتحسين قابليته للقراءة. بعد استخدام هذه الأداة/الخدمة، قام المؤلفون بمراجعة وتحرير المحتوى حسب الحاجة ويتحملون المسؤولية الكاملة عن محتوى النشر.

بيان إعلان المصالح

لا شيء.

بيان مساهمة مؤلفي CRediT

ستانيسلاف إيفانوف: التصور، المنهجية، التحقيق، التصور البصري، الإشراف، الكتابة – المسودة الأصلية، الكتابة – المراجعة والتحرير. محمد سليمان: التصور، المنهجية، التحقيق، الكتابة – المسودة الأصلية، الكتابة – المراجعة والتحرير. آرني توومي: التصور، المنهجية، التحقيق، الكتابة – المسودة الأصلية، الكتابة – المراجعة والتحرير. ناصر الحمير الكعيثري: التصور، المنهجية، التحقيق، التحليل الرسمي، التصور البصري، الكتابة – المسودة الأصلية، الكتابة – المراجعة والتحرير. الأمیر ن. العلوي: التصور، المنهجية، التحقيق، الكتابة – المسودة الأصلية، الكتابة – المراجعة والتحرير.

توفر البيانات

ستكون البيانات متاحة عند الطلب.

الملحق 1. التدابير

المتغيرات عناصر مصدر
قوة الذكاء الاصطناعي التوليدي
للسؤالين أدناه، فكر في الذكاء الاصطناعي التوليدي الذي استخدمته بشكل أساسي:
أعتقد أن تطبيق الذكاء الاصطناعي التوليدي:
STRN 1*** فهمت تفاصيل اللغة البشرية
ستران تفاعل بطريقة محادثة وإنسانية.
ستران يمكن أن يجيب على الأسئلة المتابعة،
STRN 4*** … يمكن أن تعترف بأخطائها.
STRN 5 *** … يمكن أن ترفض الطلبات غير المناسبة.
ستران يمكن أن يتتبع المحادثة الجارية.
STRN 7 … المحتوى الذي تم إنشاؤه كان مفيدًا لي
تم تطويره بواسطة المؤلفين
فوائد الذكاء الاصطناعي التوليدي
للسؤالين أدناه، فكر في الذكاء الاصطناعي التوليدي الذي استخدمته بشكل أساسي:
أعتقد أن تطبيق الذكاء الاصطناعي التوليدي:
BNFT1 … عزز كفاءة عملي
BNFT2 … حسّن جودة العمل الذي أقوم به.
BNFT3 … ساعدني في إنجاز مهامي بشكل أسرع.
بشكل عام، أجد أن استخدام تطبيق الذكاء الاصطناعي التوليدي مفيد في عملي.
هسو وآخرون [45] وتم توسيعه بواسطة المؤلفين
نقاط ضعف الذكاء الاصطناعي التوليدي
للسؤالين أدناه، فكر في الذكاء الاصطناعي التوليدي الذي استخدمته بشكل أساسي:
أعتقد أن تطبيق الذكاء الاصطناعي التوليدي:
WEAK1 … كان يولد معلومات خاطئة.
WEAK2*** … لم يكن قادرًا على التفكير الأخلاقي.
WEAK3 … افتقر إلى الموثوقية بشأن المعرفة الواقعية.
WEAK4* … واجهت صعوبة في تقديم المراجع المناسبة للمصادر التي كانت تستخدمها.
WEAK5** … لم يكن قادرًا على طرح أسئلة توضيحية عند تلقيه تعليمات غامضة.
WEAK6 … لم يكن يقدم إجابة كافية على أسئلتي.
WEAK7 … واجه صعوبة في توليد ردود على أسئلة معقدة أو مجردة
تم تطويره بواسطة المؤلفين
مخاطر الذكاء الاصطناعي التوليدي
للسؤالين أدناه، فكر في الذكاء الاصطناعي التوليدي الذي استخدمته بشكل أساسي:
أعتقد أن تطبيق الذكاء الاصطناعي التوليدي:
RSK1*** … أنتجت ردودًا قد تكون متحيزة.
RSK2*** … كانت تستخدم البيانات الحساسة التي شاركتها معها كبيانات تدريب.
RSK3*** … قد تحل محل العديد من الوظائف المعتمدة على البحث.
قد يقلل RSK4 من مصداقية عملي
قد يقلل RSK5 من ثقة الآخرين في عملي
قد يتم حظر RSK6** من قبل مؤسستي
قد يتم حظر RSK7*** من قبل المجلات الأكاديمية
تم تطويره بواسطة المؤلفين
الموقف
ATTD1- بالنسبة لي، استخدام الذكاء الاصطناعي التوليدي سيء للغاية
ATTD2- بالنسبة لي، استخدام الذكاء الاصطناعي التوليدي غير مرغوب فيه للغاية
ATTD3 – بالنسبة لي، استخدام الذكاء الاصطناعي التوليدي غير مريح للغاية
ATTD4- بالنسبة لي، استخدام الذكاء الاصطناعي التوليدي أمر غبي للغاية
ATTD5- بالنسبة لي، استخدام الذكاء الاصطناعي التوليدي غير مفضل للغاية.
ATTD6- بالنسبة لي، استخدام الذكاء الاصطناعي التوليدي غير ممتع على الإطلاق
ATTD7**- بالنسبة لي، استخدام الذكاء الاصطناعي التوليدي سلبي للغاية
[٤٤]
المتغيرات عناصر مصدر
التحكم السلوكي المدرك PRCV1- سواء استخدمت الذكاء الاصطناعي التوليدي أثناء عملي أم لا، فهذا يعتمد تمامًا عليّ. PRCV2- أنا واثق أنه إذا أردت، يمكنني استخدام الذكاء الاصطناعي التوليدي أثناء عملي. [٤٤]
PRCV3- لدي موارد ووقت وفرص لاستخدام الذكاء الاصطناعي التوليدي أثناء عملي
نية استخدام الذكاء الاصطناعي التوليدي INTN1- يستحق الأمر استخدام الذكاء الاصطناعي التوليدي أثناء القيام بعملي [٤٥]
INTN2- سأستخدم الذكاء الاصطناعي التوليدي بشكل متكرر أثناء عملي في المستقبل.
الاستخدام الفعلي للذكاء الاصطناعي التوليدي ACTU1- أستخدم الذكاء الاصطناعي التوليدي يوميًا [43]
ACTU2- أستخدم الذكاء الاصطناعي التوليدي بشكل متكرر
*** تمت إزالته من جميع النماذج بسبب انخفاض تحميل المؤشر.
** تمت إزالتها من نموذجين (العينة الكلية والطلاب) بسبب انخفاض تحميل المؤشر.
تمت إزالة نموذج الطلاب فقط بسبب انخفاض تحميل المؤشر.

الملحق 2. ملاءمة النموذج ومؤشرات الجودة

مقياس بشكل عام محاضرات الطلاب القيمة الموصى بها
معامل المسار المتوسط (APC) 0.251، P < 0.001 0.266، P < 0.001 0.255، P < 0.001
متوسط R-squared (ARS) 0.433، 0.474، P < 0.001 0.440، P < 0.001
متوسط معامل التحديد المعدل (AARS) 0.426، P < 0.001
متوسط VIF للكتلة (AVIF) 1.372 1.273 1.508 مثاليًا
متوسط عامل التضخم المتعلق بالخطية الكاملة (AFVIF) 1.989 ٢.١٠٤ ٢.١٤١ من الناحية المثالية
تيننهاوس غوف (غوف) 0.568 0.595 0.560 كبير
نسبة مفارقة سيمبسون (SPR) 0.938 0.938 1.000 مقبول إذا ، بشكل مثالي
نسبة مساهمة R-squared (RSCR) 0.998 0.996 1.000 مقبول إذا ، بشكل مثالي
نسبة الكبت الإحصائي (SSR) 1.000 0.938 0.938 مقبول إذا
نسبة اتجاه السببية الثنائية غير الخطية (NLBCDR) 0.875 0.844 0.906 مقبول إذا

References

[1] J. Wirtz, V. Pitardi, How intelligent automation, service robots, and AI will reshape service products and their delivery, Ital. J. Market. 2023 (2023) 289-300, https:// doi.org/10.1007/s43039-023-00076-1.
[2] O. Ali, P. Murray, M. Momin, F.S. Al-Anzi, The knowledge and innovation challenges of ChatGPT: a scoping review, Technol. Soc. 75 (2023) 102402, https:// doi.org/10.1016/j.techsoc.2023.102402.
[3] S.A. Bin-Nashwan, M. Sadallah, M. Bouteraa, Use of ChatGPT in academia: academic integrity hangs in the balance, Technol. Soc. 75 (2023) 102370, https:// doi.org/10.1016/j.techsoc.2023.102370.
[4] Y.K. Dwivedi, N. Kshetri, L. Hughes, E.L. Slade, A. Jeyaraj, A.K. Kar, A. M. Baabdullah, A. Koohang, V. Raghavan, M. Ahuja, “So what if ChatGPT wrote it?” Multidisciplinary perspectives on opportunities, challenges and implications of generative conversational AI for research, practice and policy, Int. J. Inf. Manag. 71 (2023) 102642, https://doi.org/10.1016/j.ijinfomgt.2023.102642.
[5] S. Rice, S.R. Crouse, S.R. Winter, C. Rice, The advantages and limitations of using ChatGPT to enhance technological research, Technol. Soc. 76 (2024) 102426, https://doi.org/10.1016/j.techsoc.2023.102426.
[6] H.S. Sætra, Generative AI: here to stay, but for good? Technol. Soc. 75 (2023) 102372 https://doi.org/10.1016/j.techsoc.2023.102372.
[7] A. Susarla, R. Gopal, J.B. Thatcher, S. Sarker, The Janus effect of generative AI: charting the path for responsible conduct of scholarly activities in information systems, Inf. Syst. Res. 34 (2) (2023) 399-408, https://doi.org/10.1287/isre.2023. ed.v34.n2.
[8] R. Vogler, 2023 – an AI university space Odyssey, Retrieved 15th February 2024 from, ROBONOMICS: J. Autom. Econ. 5 (2024) 55, https://journal.robonomics. science/index.php/rj/article/view/55.
[9] S. Ivanov, M. Soliman, Game of algorithms: ChatGPT implications for the future of tourism education and research, J. Tourism Futur. 9 (2) (2023) 214-221, https:// doi.org/10.1108/JTF-02-2023-0038.
[10] T.K. Chiu, The impact of Generative AI (GenAI) on practices, policies and research direction in education: a case of ChatGPT and Midjourney, Interact. Learn. Environ. (2023) 1-17, https://doi.org/10.1080/10494820.2023.2253861.
[11] M. Farrokhnia, S.K. Banihashem, O. Noroozi, A. Wals, A SWOT analysis of ChatGPT: implications for educational practice and research, Innovat. Educ. Teach. Int. (2023) 1-15, https://doi.org/10.1080/14703297.2023.2195846.
[12] A. Gilson, C. Safranek, T. Huang, V. Socrates, L. Chi, R.A. Taylor, D. Chartash, How well does ChatGPT do when taking the medical licensing exams? The implications of large language models for medical education and knowledge assessment, medRxiv (2022) 1-9, https://doi.org/10.1101/2022.12.23.22283901.
[13] F.M. Megahed, Y.-J. Chen, J.A. Ferris, S. Knoth, L.A. Jones-Farmer, How generative AI models such as ChatGPT can be (mis)used in SPC practice, education, and research? An exploratory study, Qual. Eng. 36 (2) (2024) 287-315, https://doi. org/10.1080/08982112.2023.2206479.
[14] S. Biswas, ChatGPT and the future of medical writing, Radiology 307 (2) (2023) e223312, https://doi.org/10.1148/radiol.223312.
[15] D.R. Cotton, P.A. Cotton, J.R. Shipway, Chatting and cheating: ensuring academic integrity in the era of ChatGPT, Innovat. Educ. Teach. Int. 61 (2) (2024) 228-239, https://doi.org/10.1080/14703297.2023.2190148.
[16] E. Kasneci, K. Seßler, S. Küchemann, M. Bannert, D. Dementieva, F. Fischer, U. Gasser, G. Groh, S. Günnemann, E. Hüllermeier, ChatGPT for good? On opportunities and challenges of large language models for education, Learn. Indiv Differ 103 (2023) 102274, https://doi.org/10.1016/j.lindif.2023.102274.
[17] A. Strzelecki, S. ElArabawy, Investigation of the moderation effect of gender and study level on the acceptance and use of generative AI by higher education students: comparative evidence from Poland and Egypt, Br. J. Educ. Technol. (2024), https://doi.org/10.1111/bjet. 13425.
[18] M. Jaboob, M. Hazaimeh, A.M. Al-Ansi, Integration of generative AI techniques and applications in student behavior and cognitive achievement in Arab higher education, Int. J. Hum. Comput. Interact. (2024) 1-14, https://doi.org/10.1080/ 10447318.2023.2300016.
[19] Y. Wang, W. Zhang, Factors influencing the adoption of generative AI for art designing among Chinese generation Z: a structural equation modeling approach, IEEE Access 11 (2023) 143272-143284, https://doi.org/10.1109/ ACCESS. 2023.3342055.
[20] I. Ajzen, The theory of planned behavior, Organ. Behav. Hum. Decis. Process. 50 (2) (1991) 179-211, https://doi.org/10.1016/0749-5978(91)90020-T.
[21] I. Ajzen, The theory of planned behavior: frequently asked questions, Hum. Behav. Emerg. Technol. 2 (4) (2020) 314-324, https://doi.org/10.1002/hbe2.195.
[22] H. Knauder, C. Koschmieder, Individualized student support in primary school teaching: a review of influencing factors using the Theory of Planned Behavior (TPB), Teach. Teach. Educ. 77 (2019) 66-76, https://doi.org/10.1016/j. tate.2018.09.012.
[23] T. Teo, C. Beng Lee, Explaining the intention to use technology among student teachers: an application of the Theory of Planned Behavior (TPB), Campus-Wide Inf. Syst. 27 (2) (2010) 60-67, https://doi.org/10.1108/10650741011033035.
[24] M. Bosnjak, I. Ajzen, P. Schmidt, The theory of planned behavior: selected recent advances and applications, Eur. J. Psychol. 16 (3) (2020) 352-356, https://doi. org/10.5964/ejop.v16i3.3107.
[25] M. Conner, C.J. Armitage, Extending the theory of planned behavior: a review and avenues for further research, J. Appl. Soc. Psychol. 28 (15) (1998) 1429-1464, https://doi.org/10.1111/j.1559-1816.1998.tb01685.x.
[26] Y. Wang, C. Dong, X. Zhang, Improving MOOC learning performance in China: an analysis of factors from the TAM and TPB, Comput. Appl. Eng. Educ. 28 (6) (2020) 1421-1433, https://doi.org/10.1002/cae.22310.
[27] J. Cheon, S. Lee, S.M. Crooks, J. Song, An investigation of mobile learning readiness in higher education based on the theory of planned behavior, Comput. Educ. 59 (3) (2012) 1054-1064, https://doi.org/10.1016/j.compedu.2012.04.015.
[28] K.M. White, I. Thomas, K.L. Johnston, M.K. Hyde, Predicting attendance at peerassisted study sessions for statistics: role identity and the theory of planned behavior, J. Soc. Psychol. 148 (4) (2008) 473-492, https://doi.org/10.3200/ SOCP.148.4.473-492.
[29] D. Baidoo-Anu, L.O. Ansah, Education in the era of generative artificial intelligence (AI): understanding the potential benefits of ChatGPT in promoting teaching and learning, J. AI 7 (1) (2023) 52-62, https://doi.org/10.61969/jai.1337500.
[30] J. Qadir, Engineering education in the era of ChatGPT: promise and pitfalls of generative AI for education, in: 2023 IEEE Global Engineering Education Conference (EDUCON), Kuwait, Kuwait, 2023, pp. 1-9, https://doi.org/10.1109/ EDUCON54358.2023.10125121.
[31] Stanford, Responsible AI at Stanford, 2023. Retrieved 16th December 2023 from, https://uit.stanford.edu/security/responsibleai.
[32] UNESCO, Generative Artificial Intelligence in Education: what Are the Opportunities and Challenges, 2023. Retrieved 16th December 2023 from, https:// www.unesco.org/en/articles/generative-artificial-intelligence-education-what-are -opportunities-and-challenges.
[33] J. Walmsley, Artificial intelligence and the value of transparency, AI Soc. 36 (2) (2021) 585-595, https://doi.org/10.1007/s00146-020-01066-z.
[34] S. Noy, W. Zhang, Experimental evidence on the productivity effects of generative artificial intelligence, SSRN 4375283, https://doi.org/10.2139/ssrn.4375283, 2023.
[35] M. Shanahan, K. McDonell, L. Reynolds, Role play with large language models, Nature 623 (2023) 493-498, https://doi.org/10.1038/s41586-023-06647-8.
[36] L. Kohnke, B.L. Moorhouse, D. Zou, ChatGPT for language teaching and learning, 00336882231162868, RELC J. (2023).
[37] I. Carvalho, S. Ivanov, ChatGPT for tourism: applications, benefits and risks, Tour. Rev. 79 (2) (2024) 290-303, https://doi.org/10.1108/TR-02-2023-0088.
[38] E. Bender, T. Gebru, A. McMillan-Major, S. Shmitchell, On the dangers of stochastic parrots: can language models be too big?, in: FAccT ’21: Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, 2021, pp. 610-623, https://doi.org/10.1145/3442188.3445922.
[39] J. Baker-Brunnbauer, TAII framework for trustworthy AI systems, Retrieved 23rd June 2023 from, ROBONOMICS: J. Autom. Econ. 2 (2021) 17, https://journal. robonomics.science/index.php/rj/article/view/17.
[40] Z. Li, The dark side of chatgpt: legal and ethical challenges from stochastic parrots and hallucination, arXiv preprint arXiv:2304.14347, https://doi.org/10.4855 0/arXiv.2304.14347, 2023.
[41] S. Ivanov, The dark side of artificial intelligence in higher education, Serv. Ind. J. 43 (15-16) (2023) 1055-1082, https://doi.org/10.1080/ 02642069.2023 .2258799.
[42] J.F. Hair, C.M. Ringle, M. Sarstedt, PLS-SEM: indeed a silver bullet, J. Market. Theor. Pract. 19 (2) (2011) 139-152, https://doi.org/10.2753/MTP10696679190202.
[43] W.H. DeLone, E.R. McLean, The DeLone and McLean model of information systems success: a ten-year update, J. Manag. Inf. Syst. 19 (4) (2003) 9-30, https://doi.org/ 10.1080/07421222.2003.11045748.
[44] H. Han, L.-T.J. Hsu, C. Sheu, Application of the theory of planned behavior to green hotel choice: testing the effect of environmental friendly activities, Tourism Manag. 31 (3) (2010) 325-334, https://doi.org/10.1016/j.tourman.2009.03.013.
[45] C.-L. Hsu, H.-P. Lu, H.-H. Hsu, Adoption of the mobile Internet: an empirical study of multimedia message service (MMS), Omega 35 (6) (2007) 715-726, https://doi. org/10.1016/j.omega.2006.03.005.
[46] N. Kock, WarpPLS User Manual: Version 8.0 Scriptwarp Systems: Laredo TX, USA, 2022. Retrieved 16th December 2023 from, https://scriptwarp.com/warppls/U serManual_v_8_0.pdf.
[47] N. Kock, L. Gaskins, The mediating role of voice and accountability in the relationship between Internet diffusion and government corruption in Latin America and Sub-Saharan Africa, Inf. Technol. Dev. 20 (1) (2014) 23-43, https:// doi.org/10.1080/02681102.2013.832129.
[48] J. Henseler, G. Hubona, P.A. Ray, Using PLS path modeling in new technology research: updated guidelines, Ind. Manag. Data Syst. 116 (1) (2016) 2-20, https:// doi.org/10.1108/IMDS-09-2015-0382.
[49] M. Soliman, S. Ivanov, C. Webster, The psychological impacts of COVID-19 outbreak on research productivity: a comparative study of tourism and nontourism scholars, J. Tour. Dev. 35 (2021) 23-52, https://doi.org/10.34624/rtd. v0i35.24616.
[50] M. Soliman, R. Sinha, F. Di Virgilio, M.J. Sousa, R. Figueiredo, Emotional intelligence outcomes in higher education institutions: empirical evidence from a Western context, 00332941231197165, Psychol. Rep. (2023), https://doi.org/ 10.1177/00332941231197165.
[51] M.A.Q. Tran, T. Vo-Thanh, M. Soliman, B. Khoury, N.N.T. Chau, Self-compassion, mindfulness, stress, and self-esteem among Vietnamese university students: psychological well-being and positive emotion as mediators, Mindfulness 13 (10) (2022) 2574-2586, https://doi.org/10.1007/s12671-022-01980-x.
[52] J.F. Hair, M.C. Howard, C. Nitzl, Assessing measurement model quality in PLS-SEM using confirmatory composite analysis, J. Bus. Res. 109 (2020) 101-110, https:// doi.org/10.1016/j.jbusres.2019.11.069.
[53] C. Fornell, D.F. Larcker, Structural equation models with unobservable variables and measurement error: Algebra and statistics, J. Market. Res. 18 (3) (1981) 382-388, https://doi.org/10.1177/002224378101800313.
[54] J. Henseler, C.M. Ringle, M. Sarstedt, A new criterion for assessing discriminant validity in variance-based structural equation modeling, J. Acad. Market. Sci. 43 (2015) 115-135, https://doi.org/10.1007/s11747-014-0403-8.
[55] P.M. Podsakoff, S.B. MacKenzie, J.-Y. Lee, N.P. Podsakoff, Common method biases in behavioral research: a critical review of the literature and recommended remedies, J. Appl. Psychol. 88 (5) (2003) 879, https://doi.org/10.1037/00219010.88.5.879.
[56] F. Kock, A. Berbekova, A.G. Assaf, Understanding and managing the threat of common method bias: detection, prevention and control, Tourism Manag. 86 (2021) 104330, https://doi.org/10.1016/j.tourman.2021.104330.
[57] T. Gundu, Chatbots: a framework for improving information security behaviours using ChatGPT, in: S. Furnell, N. Clarke (Eds.), Human Aspects of Information Security and Assurance, IFIP Advances in Information and Communication Technology, vol. 674, Springer, Cham, 2023, pp. 418-431, https://doi.org/ 10.1007/978-3-031-38530-8_33.
[58] C.S. Shah, S. Mathur, S.K. Vishnoi, Continuance intention of ChatGPT use by students, TDIT 2023, in: S.K. Sharma, Y.K. Dwivedi, B. Metri, B. Lal, A. Elbanna (Eds.), Transfer, Diffusion and Adoption of Next-Generation Digital Technologies, IFIP Advances in Information and Communication Technology, Springer, Cham, 2024, pp. 159-175, https://doi.org/10.1007/978-3-031-50188-3_14.
[59] A.M. Al-Zahrani, The impact of generative AI tools on researchers and research: implications for academia in higher education, Innovat. Educ. Teach. Int. (2023) 1-15, https://doi.org/10.1080/14703297.2023.2271445.
[60] V. Dubljević, Colleges and universities are important stakeholders for regulating large language models and other emerging AI, Technol. Soc. 76 (2024) 102480, https://doi.org/10.1016/j.techsoc.2024.102480.
[61] E.M. Rogers, Diffusion of Innovations, 1983, third ed., The Free Press, London, 1962.
[62] V. Venkatesh, F.D. Davis, A theoretical extension of the technology acceptance model: four longitudinal field studies, Manag. Sci. 46 (2) (2000) 186-204, https:// doi.org/10.1287/mnsc.46.2.186.11926.
[63] V. Venkatesh, M.G. Morris, G.B. Davis, F.D. Davis, User acceptance of information technology: toward a unified view, MIS Q. (2003) 425-478, https://doi.org/ .
Dr. Stanislav Ivanov is currently Professor and Vice-Rector (Research) at Varna University of Management, Bulgaria (http://www.vum.bg) and Director of Zangador Research Institute (https://www.zangador.institute/en/). Prof. Ivanov is the Founder and Editor-in-chief of two academic journals: European Journal of Tourism Research (http://ejtr.vumk.eu) and ROBONOMICS: The Journal of the Automated Economy (https ://journal.robonomics.science). His research interests include robonomics, robots in tourism/hospitality, the economics of technology, etc. For more information about Prof. Ivanov please visit his personal website: http://www.stanislavivanov.com.
Dr. Mohammad Soliman is currently the Head of Research & Consultation Department at the University of Technology and Applied Sciences, Salalah, Oman. He is also a Full Professor at the Faculty of Tourism & Hotels, Fayoum University, Egypt. He has published multiple papers in high-rank journals indexed in WoS and Scopus. Additionally, he sits on the editorial board of different academic journals and serves as a reviewer for several top-tier journals. He has successfully supervised and examined several master’s and PhD theses. His research interests include tourism marketing, consumer behaviour, branding, AI-enabled education and marketing, tourism management, PLS-SEM, bibliometrics, and literature review studies.
Dr. Aarni Tuomi is Senior Lecturer at Haaga-Helia University of Applied Sciences. His research, teaching and consultancy projects explore the intersection of emerging technologies and service business. His research has explored e.g. service robotics, artificial intelligence, digital platforms and food technology, as well as experience design and innovation. His work is regularly featured in industry trade magazines and his research has been published in top academic journals, e.g. Annals of Tourism Research, Psychology & Marketing.
Dr. Nasser Alhamar Alkathiri is an Assistant Professor and the Deputy Assistant Vice Chancellor for Postgraduate Studies, Scientific Research and Innovation, University of Technology and Applied Sciences, Salalah, Oman. Dr. Nasser holds a PhD in Knowledge Transfer from Plymouth University, the UK. He also holds a Master degree in International Business from Sydney University, Australia. His research interests include knowledge management, knowledge transfer, international, business, entrepreneurship and staff localization. He has published multiple papers in reputed journals indexed in WoS and Scopus (e.g. Journal of Knowledge Management, International Journal of Finance & Economics and International Journal of Contemporary Hospitality Management). He has several participations in local and international conferences.
Dr. Al-Amir obtained his bachelor’s degree from the Sultanate of Oman and pursued his master’s and doctoral degrees in the United Kingdom. Over the course of 23 years, he held various positions. He currently holds the position of Assistant Vice Chancellor of the University of Technology and Applied Sciences in Ibri. Dr. Al-Amir’s area of interest is finance, entrepreneurship, tourism and artificial intelligence. He succeeded in working on two research projects funded by the Scientific Research Council of Oman.

    • Corresponding author. Varna University of Management, 13A Oborishte Str., 9000 Varna, Bulgaria.
    E-mail addresses: stanislav.ivanov@vumk.eu, info@zangador.institute (S. Ivanov), msoliman.sal@cas.edu.om (M. Soliman), Aarni.Tuomi@haaga-helia.fi (A. Tuomi), nasser2014.sal@cas.edu.om (N.A. Alkathiri), alameer.alalawi@utas.edu.om (A.N. Al-Alawi).
    web: http://stanislavivanov.com/.

Journal: Technology in Society, Volume: 77
DOI: https://doi.org/10.1016/j.techsoc.2024.102521
Publication Date: 2024-03-25

HUOM! Tämä on alkuperäisen artikkelin rinnakkaistallenne. Rinnakkaistallenne saattaa erota alkuperäisestä sivutukseltaan ja painoasultaan.
PLEASE NOTE! This in an electronic self-archived version of the original article. This reprint may differ from the original in pagination and typographic detail.
Viittaa alkuperäiseen lähteeseen:
Cite the final publication:
Ivanov, S., Soliman, M., Tuomi, A., Alkathiri, N. A. & Al-Alawi, A. N. (2024) Drivers of generative Al adoption in higher education through the lens of the Theory of Planned Behaviour. Technology in Society, 77(June 2024), 102521. https://doi.org/10.1016/j.techsoc.2024.102521
© 2024 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Drivers of generative AI adoption in higher education through the lens of the Theory of Planned Behaviour

Stanislav Ivanov , Mohammad Soliman , Aarni Tuomi , Nasser Alhamar Alkathiri , Alamir N. Al-Alawi Varna University of Management, 13A Oborishte Str., 9000 Varna, Bulgaria Zangador Research Institute, 9010 Varna, Bulgaria Research and Consultation Department, University of Technology and Applied Sciences, Salalah, Oman & Faculty of Tourism & Hotels, Fayoum University, Fayoum, Egypt Haaga-Helia University of Applied Sciences, Helsinki, Finland Business Administration Department, College of Economics and Business Administration, University of Technology and Applied Sciences-Salalah, Salalah, Oman University of Technology and Applied Sciences, Ibri, Oman

ARTICLE INFO

Keywords:

Generative AI
Theory of planned behaviour
Higher education

Abstract

Drawing on the Theory of Planned Behaviour (TPB), this study investigates the relationship between the perceived benefits, strengths, weaknesses, and risks of generative AI (GenAI) tools and the fundamental factors of the TPB model (i.e., attitude, subjective norms, and perceived behavioural control). The study also investigates the structural association between the TPB variables and intention to use GenAI tools, and how the latter might affect the actual usage of GenAI tools in higher education. The paper adopts a quantitative approach, relying on an anonymous self-administered online questionnaire to gather primary data from 130 lecturers and 168 students in higher education institutions (HEIs) in several countries, and PLS-SEM for data analysis. The results indicate that although lecturers’ and students’ perceptions of the risks and weaknesses of GenAI tools differ, the perceived strengths and advantages of GenAI technologies have a significant and positive impact on their attitudes, subjective norms, and perceived behavioural control. The TPB core variables positively and significantly impact lecturers’ and students’ intentions to use GenAI tools, which in turn significantly and positively impact their adoption of such tools. This paper advances theory by outlining the factors shaping the adoption of GenAI technologies in HEIs. It provides stakeholders with a variety of managerial and policy implications for how to formulate suitable rules and regulations to utilise the advantages of these tools while mitigating the impacts of their disadvantages. Limitations and future research opportunities are also outlined.

1. Introduction

The revolution in the service industry, including the educational sector, began when technology rapidly advanced in terms of intelligence and power while becoming more compact, lightweight, and affordable. This covers both hardware like smart self-service technologies and software and systems like machine learning (ML) and generative artificial intelligence (GenAI) tools [1]. GenAI is a term used to describe a class of AI models producing ostensibly novel output such as text, images, video, music, or other types of media. While GenAI approaches have been available for a while, the launch of ChatGPT sparked a flood
of discussions in the media, online forums, and academic communities [2-8]. As a result, researchers and practitioners are becoming increasingly interested in the implications of GenAI applications, especially those based on Large Language Models (LLMs), on human learning, knowledge generation, and the nature of employment in the coming years [4]. Ivanov and Soliman [9] indicated that, in the long run, LLM-based chatbots would revolutionise research and education. If adopted successfully, they could be used as online instructors, curriculum developers, markers, and contributors to scholarly publications. LLMs would also be essential in rethinking education from “teacher-student” interactions to “teacher-AI-student” co-creation [9], shifting
the emphasis of lecturers to developing novel tasks and activities with GenAI applications. This justifies the growing number of recent publications about the advantages and drawbacks of using GenAI technologies, such as ChatGPT, in research and education (e.g., Ref. [9-12]).
Considering the substantial opportunities for using GenAI, there is a crucial need to shift academic focus from lamenting the collapse of education and research to considering how students and researchers will and should use such tools [7]. According to Megahed et al. [13], the potential of GenAI models to provide code, explain fundamental concepts, and generate knowledge might revolutionise statistical process control practice, teaching, and research. These technologies, however, are still in the early phases of deployment and are susceptible to misuse and misunderstanding. Therefore, a thorough empirical analysis is needed to provide an in-depth overview and comprehension of the potential of using such applications for educational and research purposes. Prior studies provided valuable insights into the use of GenAI tools in different settings including the educational and research context through the lenses of different theories (e.g., Ref. [9,12,14-16]). For instance, using the Unified Theory of Acceptance and Use of Technology (UTAUT), Strzelecki and ElArabawy [17] demonstrated how social influence, effort expectancy, and performance expectancy have a major impact on behavioural intention. The actual ChatGPT-using behaviour of Egyptian and Polish university students was influenced by behavioural intention when considered along with facilitating conditions. Jaboob et al. [18] investigated how university students’ cognitive achievement was affected by GenAI tools and applications in three Arab nations: Yemen, Jordan, and Oman. The findings showed that the cognitive achievement of students at Arab HEIs was positively and significantly impacted by GenAI approaches and applications. Additionally, the results demonstrated that student behaviour improved the association between GenAI tools and cognitive achievement. Drawing on the UTAUT2 model and the Technology Readiness Index, Wang and Zhang’s [19] study assessed the elements and personal traits that motivated Generation Z to adopt GenAI-assisted design and found that the intention to use GenAI was positively influenced by effort expectancy, price value, and hedonic motivation. Performance expectancy, effort expectancy, price value, and hedonic motivation were all strongly influenced by optimism and innovativeness. Optimism and the intention to use GenAI were significantly influenced by trait curiosity. Despite these studies, there is insufficient research investigating how the positive and negative aspects of generative AI tools can effectively predict the core constructs of the Theory of Planned Behaviour (TPB), such as attitudes, perceived behavioural control, subjective norms, and behavioural intentions. Consequently, the current article addresses this research gap by extending the TPB model to the context of generative AI.
The present study seeks to unveil the key drivers of GenAI adoption in higher education. More specifically, this paper aims to (1) examine the impact of strengths, benefits, weaknesses, and risks of using GenAI on attitudes, subjective norms, and perceived behaviour control (PBC) of students and lecturers; (2) investigate the impact of attitudes, subjective norms, and PBC of students and lecturers on their intention to use GenAI applications; and (3) test the connection between students and lecturers’ intention and their actual usage of GenAI applications in their study and/or research.
In doing so, this paper explores both the advantages and disadvantages that using GenAI technologies in research and educational settings may bring about. As a result, concerned stakeholders (e.g., senior management and educators at higher education institutions) may design better guidelines and policies to utilise the advantages of GenAI tools while minimising any potential negatives by developing a thorough grasp of the implications and potential difficulties of incorporating them in research and education. Additionally, this research aids in improving the understanding of human behaviour particularly in relation to human-computer interaction, making it easier to create interventions and regulations that are more specifically focused on encouraging desired behavioural outcomes in education and research.
This paper is organised as follows. The next section provides the literature review and develops the hypotheses, while the third elaborates on the methodology. The fourth section presents the results, the fifth section discusses the theoretical and managerial implications, and the last section identifies research limitations and future research directions.

2. Literature review and hypotheses development

2.1. Theory of planned behaviour

The Theory of Planned Behaviour, first put forward by Ajzen in the late 1980s [20,21], offers a useful lens through which to explore the dynamics underlying human behaviour in the context of teaching and learning, both in general and in the context of technology-use as part of education and research [22,23]. Generally, TPB posits that behaviour is shaped by attitudes, subjective norms, and perceived behavioural control, which collectively influence the formation of behavioural intentions and subsequent actions [24]. Central to the TPB is the construct of attitudes, which captures evaluative judgments of various behaviours. In the context of teaching and learning, positive attitudes, rooted in a favourable perception of the outcomes associated with teaching and learning activities, according to TPB drive the development of intentions and consequent instructional engagement. The subjective norm examines the influence of societal and peer expectations on behavioural intention and actual behaviour [25]. The perceived approval or disapproval of fellow students, colleagues, administrators, and the broader community around higher education acts as a potent motivational factor, steering teachers and learners towards alignment with perceived educational norms.
TPB also introduces the notion of perceived behavioural control, which in the context of teaching and learning refers to encompassing beliefs in an ability to effectively execute instructional strategies. Greater perceived control is anticipated to bolster both the intention to engage in specific teaching practices and the subsequent translation of intentions into actual classroom actions [22]. These intertwined components collectively contribute to the shaping of behavioural intentions, which serve as crucial antecedents to actual behaviours.
TPB’s versatility finds resonance within the context of teaching and learning, as underscored by its widespread adoption in educational research to study, e.g., the usefulness of massive open online courses [26] or mobile learning [27]. Its applications have spanned diverse facets of pedagogy, ranging from the integration of innovative teaching technologies to the adoption of student-centred educational approaches such as peer learning [28]. In educational psychology, TPB has been used to investigate instructors’ adoption of evidence-based teaching practices, assessment strategies, and classroom management techniques [22].
By dissecting the connections between attitudes, subjective norms, and perceived control, TPB offers a framework for understanding educational decision-making. Its application underscores its potential not only in explicating teachers’ and students’ behaviours but also in guiding efforts to design targeted interventions, e.g., the use of generative AI as part of teaching and learning.

2.2. Generative AI in teaching and learning

GenAI has emerged as a transformative technology with multifaceted implications for education [2,4]. Within the domain of teaching and learning, GenAI has garnered increasing attention due to its potential to reshape pedagogical approaches and learning experiences in a myriad of fields, e.g., social sciences, mathematics, and engineering [9,29,30]. To address the generative AI elephant in the classroom, both individual educational institutions as well as multinational education organisations such as UNESCO have joined the discussion on how GenAI should best be used in teaching and learning [31,32]. Most educators and
researchers seem to conclude that the advent of GenAI presents a double-edged sword, whereby on one hand, the integration of GenAI in education offers a range of potential benefits both for the teacher and the learner, but on the other hand, it also brings forward new challenges and potential for misuse [15,16].
In terms of benefits, Dwivedi et al. [4] highlighted that personalised learning experiences, a cornerstone of contemporary educational philosophy, might be significantly enhanced through the capabilities of generative AI. The technology enables the creation of ultra-tailored learning materials, assessments, and feedback mechanisms, thereby catering to individual student’s contextual needs, special requirements and learning preferences. This personalisation and the availability of a private tutor may foster greater engagement and a deeper understanding of the subject matter [10]. However, some students might find AI tutors easier to accept than others, while the data on which the AI has been trained to personalise learning material should also be under open scrutiny to improve what Walmsley [33] calls ‘functional transparency’.
For the educator, generative AI presents opportunities to streamline and optimise instructional material development [10]. Educators often invest substantial time and effort in crafting learning resources such as quizzes, tutorials, reading lists and study guides. Generative AI can alleviate this burden by automating the generation of such materials, allowing educators to allocate more time to direct interactions with students [29]. This efficiency in content creation holds the potential to expedite the educational process while maintaining the quality of instruction, whereby the role of researcher/educator might start to move from creator to curator of knowledge, analogous to a shift from being an author to being an editor of a scientific publication [34].
As an example of the benefits of GenAI-based personalised learning, language learning, a complex cognitive endeavour, can benefit from generative AI’s capacity to simulate real-world language interactions (e. g., interactive chat). Language learners often struggle to find opportunities for immersive language practice. Generative AI, particularly textgenerative large language models such as GPT-4 or LLAMA2, can bridge this gap by generating lifelike conversational scenarios, providing learners with a dynamically adaptive and personalised platform to refine their linguistic skills in authentic contexts [16]. This immersive language practice – effectively a novel form of human-AI role play [35] may contribute to enhanced proficiency and confidence in communication [36].
Besides personalised educational experiences and text-editing skills (e.g., grammar check, proofreading, making arguments more concise, iterative reasoning), GenAI tools (e.g. text-to-image, text-to-video, audio-to-animation) offer opportunities for new types of creative expression within educational contexts, whereby students’ creative endeavours can be catalysed by AI-powered tools that assist in generating novel and imaginative content [37]. This augmentation of creative capabilities not only broadens the horizons of education but may also contribute to the development of problem-solving and critical thinking skills, provided that educators possess the necessary skills to proactively and confidently introduce new learning technologies to the classroom [10].
Despite many potential benefits, the integration of GenAI in higher education is not without its challenges and ethical considerations. For example, bias in AI-generated content is of paramount concern, whereby AI models can perpetuate biases present in their training dataset, potentially reinforcing stereotypes and marginalising certain demographic groups [4]. Safeguarding against bias and ensuring fairness in AI-generated educational content demands scrutiny and mitigation strategies. It also requires robust approaches to selecting and acquiring training data, whereby the efficacy of large language models is highly dependent on the availability of high-quality and balanced datasets for model training. As the output of a GenAI system is a direct result of its training, there is a pressing need for frameworks for auditing the training process and training data [38] to develop trustworthy AI systems [39].
The question of authenticity and originality, part of a broader discourse on intellectual property law about AI-generated content, requires addressing [37]. Who should own rights to AI-generated content? Who is responsible for AI-hallucinated content [40]? Under what conditions user-generated data can and cannot be used for the subsequent training of new GenAI models? These questions are yet to receive a definitive answer. In the context of higher education, one of the primary concerns is the advent of new types of plagiarism and academic integrity [16]. Striking a balance between AI assistance and the cultivation of students’ independent critical thinking skills, motivation and learning effort is a complex matter that requires nuanced approaches and pedagogical innovation. Moreover, the pedagogical efficacy of GenAI tools also warrants thorough evaluation. While the technology holds promise in enhancing learning experiences, its alignment with established pedagogical principles must be rigorously examined [36]. Educators play an important role in determining the appropriate contexts for the implementation of AI-generated content and in ensuring that such content effectively contributes to educational objectives [10], as defined in learning outcomes (micro-level) and curricula (macro-level). The tendency for LLMs to hallucinate [40] also requires educators and learners to develop skills related to critical thinking, media criticism and fact-checking.
GenAI’s integration into teaching and learning represents a significant advancement with potential benefits that span from personalised learning experiences to streamlined content creation and creative augmentation [4,9]. However, these potential advantages are accompanied by challenges related most notably to data protection, bias, representativeness and auditability of training data and training process, the authenticity of AI-generated content vis-à-vis new forms of plagiarism, and the pedagogical alignment of teacher-AI-student interactions, e.g. placing a greater emphasis on critically interrogating the model output [16]. GenAI applications may provide wrong information and invent facts, and the overreliance on AI applications to curate content over creating it may lead to the deskilling of students [41]. As educational institutions and stakeholders navigate the incorporation of GenAI, a judicious and ethically informed approach is imperative [15,32].

2.3. Hypothesis development

The extant literature indicates that GenAI offers various pedagogical benefits, including but not limited to personalised learning experiences and efficient content creation [4,29]. These perceived advantages are hypothesised to positively shape educators’ attitudes toward the adoption of GenAI technologies, as positive attitudes are frequently grounded in favourable perceptions of outcomes [20,24]. Concurrently, the perceived strengths of GenAI are likely to influence subjective norms by aligning with societal and institutional expectations for innovative teaching methods, thereby serving as a motivational factor for adoption [25,26]. Furthermore, the efficiencies gained through GenAI are anticipated to bolster perceived control over instructional and learning strategies, aligning with TPB’s emphasis on the role of perceived behavioural control in intention formation and eventual behaviour [22]. Therefore, the following hypotheses are put forward.
H1a. Perceived strengths of generative AI have a positive effect on attitude towards using generative AI.
H1b. Perceived strengths of generative AI have a positive effect on subjective norms.
H1c. Perceived strengths of generative AI have a positive effect on perceived behavioural control.
Existing research identifies a myriad of potential benefits of GenAI, such as personalised learning experiences and streamlined instructional material development [9,29]. According to TPB, these perceived benefits are posited to positively influence attitudes towards GenAI, as attitudes are commonly linked to an individual’s favourable or
unfavourable evaluations of behavioural outcomes [24]. Similarly, if GenAI’s benefits align with broader educational or societal expectations, they are likely to positively affect subjective norms, thereby serving as a motivational driver for technology adoption [25]. Lastly, the anticipated benefits of GenAI, such as the automation of labour-intensive tasks, could enhance perceived behavioural control by bolstering beliefs in the ability to successfully implement this technology in educational settings [22]. Based on this, we hypothesise that.
H2a. Perceived benefits of generative AI have a positive effect on attitude towards using generative AI.
H2b. Perceived benefits of generative AI have a positive effect on subjective norms.
H2c. Perceived benefits of generative AI have a positive effect on perceived behavioural control.
Besides strengths and benefits, existing literature highlights various challenges and ethical considerations related to the deployment of GenAI in educational settings, including issues of data protection, bias, authenticity, and pedagogical alignment [16]. TPB suggests that unfavourable perceptions of behavioural outcomes are likely to negatively impact attitudes towards a particular behaviour [20]. Similarly, if the perceived weaknesses of GenAI do not align with societal or educational expectations, this incongruence is expected to exert a negative influence on subjective norms [25]. Additionally, perceived weaknesses such as deskilling or reinforcement of biases could compromise the sense of perceived behavioural control over successful technology implementation, as suggested by TPB’s focus on perceived behavioural control as a determinant of intentions and actions [22]. The following hypotheses are thus put forward.
H3a. Perceived weaknesses of generative AI have a negative effect on attitude towards using generative AI.
H3b. Perceived weaknesses of generative AI have a negative effect on subjective norms.
H3c. Perceived weaknesses of generative AI have a negative effect on perceived behavioural control.
Several risks related to the adoption of GenAI in education, such as bias perpetuation, ethical concerns around plagiarism, and questions of academic integrity, have also been raised . According to TPB, attitudes are influenced by the evaluation of behavioural outcomes, and the risks associated with GenAI could engender negative attitudes towards its use [20]. Likewise, if societal and educational expectations are risk-averse, it is reasonable to presume that perceived risks will negatively impact subjective norms [25]. Lastly, TPB stipulates that perceived behavioural control is informed by beliefs in one’s ability to perform a behaviour successfully; therefore, perceived risks may undermine this sense of control, diminishing the likelihood of GenAI implementation in educational contexts [22]. Thus, we hypothesise that:
H4a. Perceived risks of generative AI have a negative effect on attitude towards using generative AI.
H4b. Perceived risks of generative AI have a negative effect on subjective norms.
H4c. Perceived risks of generative AI have a negative effect on perceived behavioural control.
Overall, TPB posits that attitude, subjective norms, and perceived control are antecedent variables that influence behavioural intention, which in turn leads to actual behaviour [20]. H5 proposes that a favourable attitude towards using GenAI will positively impact the intention to use it, a link that has been empirically established in multiple contexts within TPB research [23]. Similarly, H6 stipulates that societal and peer influences, represented by the subjective norms construct, positively affect behavioural intention, consistent with the
extant literature on TPB [25]. H7 extends this by positing that greater perceived behavioural control, which reflects beliefs in one’s capability to execute a behaviour, will also positively influence the intention to deploy GenAI in educational settings [22]. Finally, H8 concludes the behavioural chain by suggesting that intention, as influenced by the aforementioned variables, will positively affect actual usage, which is a fundamental tenet of TPB. Thus, we hypothesise that.
H5. Attitude towards using generative AI has a positive effect on the intention to use generative AI.
H6. Subjective norms regarding generative AI have a positive effect on the intention to use generative AI.
H7. Perceived behavioural control regarding generative AI has a positive effect on the intention to use generative AI.
H8. Intention to use generative AI has a positive effect on the actual use of generative AI.
Collectively, these hypotheses are congruent with the foundational principles of TPB and are also substantiated by the specific challenges and opportunities posed by the integration of GenAI in educational contexts. Fig. 1 presents the conceptual model of this research.

3. Methodology

3.1. Sampling and data collection procedures

Data were collected between April and June 2023 with an anonymous online questionnaire developed on Google Forms. This period enabled securing the minimum sample size. Non-probability sampling was employed as the total population size was not known since the current study targeted lecturers and students around the world who had used GenAI applications. To enhance the response rate and to avoid nonresponse bias the authors utilised three sampling techniques. First, based on the convenience sampling technique, the link to the questionnaire was disseminated to participants through their personal emails. Then, self-selection sampling was applied by posting the link on social media platforms. Finally, based on the snowball sampling technique the authors shared the link with known individuals in various countries requesting them to share the link with their networks. In total, 543 respondents completed the questionnaire of whom 240 were excluded as they declared that they did not use GenAI tools. Out of the remaining 303 responses, 5 were removed as they were not complete. Consequently, 298 respondents ( 130 lecturers and 168 students) from 47 countries were used for further analysis. The use of an international sample contributed to the diversity of respondents and the potential generalisability of results.
As per prior work, it is important to note that the most popular method for determining the number of participants for Partial Least Squares (PLS) prediction is to base the sample size on the number of regressions inside the study framework (Barclay, Higgins, & Thompson, 1995). As per the ten-times rule, the minimum sample size has to be more than ten times the largest paths directed to a variable in the model with a power level of 0.8 and significance level of 0.05 [42]. According to the current study model, the required sample size is 40 meaning that the collected responses for both groups are sufficient for PLS-SEM analysis. The demographic characteristics of the two samples are presented in Table 1.

3.2. Questionnaire design and measures

The questionnaire had three sections. The first section included a filter question that distinguished participants who had already used GenAI applications (e.g., Have you used a generative AI application in your research/studies? Yes/No). Participants who had not used these tools were not able to fill in the remaining sections. The second section
Fig. 1. Conceptual model.
included the statements of the seven variables of this study (measured on a 5-point level of agreement scale) namely: strengths of GenAI applications, benefits of GenAI applications, weaknesses of GenAI applications, risks of GenAI applications, attitude, perceived behavioural control, subjective norms, intention to use GenAI applications, actual use of GenAI applications. Appendix 1 presents the statements and their sources [43-45]. The section included two general questions as well: names of used GenAI applications and frequency of use of GenAI by usage directions (measured from 1 -never to 5 -very often). The last section covered the demographic data.

3.3. Data analysis

SEM is a statistical technique that adopts a confirmatory (i.e., testing hypotheses) strategy for the examination of a structural theory on a specific occurrence, as a means of investigating connected relationships in a complex model. Two statistical approaches are included in SEM: variance-based SEM (PLS-SEM) and covariance-based SEM (CB-SEM). CB-SEM is a technique of SEM that typically verifies or disproves the suggested hypotheses using programmes like AMOS, EQS, LISREL, and MPlus. However, many business researchers prefer to employ PLS-SEM since normal distribution may not be obtained in practice, and the CBSEM technique demands a large sample size.
PLS-SEM utilising Warp PLS V8 [46] has been used in this research for the following reasons. First, PLS-SEM is a useful technique for evaluating complicated models. The method is deemed appropriate as it is suitable for complicated models such as the one of the current study that has nine variables with multiple directions that can be tested simultaneously [47], it is recommended for estimating behavioural variables [48], and is widely applied in the higher education setting [49-51]. Moreover, unlike covariance-based methods, PLS-SEM can address not normal data distribution, a feature that is crucial for the present study. Additionally, it enables an explanation of the variation between the study constructs. Two steps in PLS-SEM were carried out, namely the measurement model and structural model in which the former deals with constructs’ reliability and validity whereas the latter measures the associations between the variables [46].
Based on WarpPLS, a series of tests have been conducted to assess the model fit of the three structural equation models, which are provided in
Appendix 2. All these assessments were clearly within the allowed range of values, indicating that the structural equation models matched their data satisfactorily.

4. Results

4.1. Measurements models

In PLS-SEM, the first step is to assess the outer model (also called ‘measurement model’) which aims to check the reliability and validity of the proposed model before examining the inner model (also named ‘structural model’) that covers the testing of suggested hypotheses. To pass this step two reliability conditions and two validity conditions should be met. The former includes indicator reliability and internal consistency reliability. In this regard, as shown in Table 2, all items with loading less than 0.7 were eliminated as suggested by Hair et al. [52]. This means that the indicator reliability has been attained, enabling the verification of further measures. Then, the model was run again with all items with loadings above this threshold. Next, the internal consistency was checked to ensure that respondents had an equal understanding of the given scales. This is done through testing the composite reliability and Cronbach’s alpha in which all values were above 0.7 , confirming that the adopted items measured their respective constructs (see Table 2).
As for the validity, convergent validity and discriminant validity were examined to ensure the validity of the model. The AVE was satisfactory for all constructs because it had values above the recommended value of 0.5 confirming the convergent validity [52]. This suggests that the items for each variable are capable of clarifying over fifty per cent of the variation in the variable they relate to. Tables present the results of the discriminant validity via two tests: Fornell and Larcker [53] and heterotrait-monotrait (HTMT) [54] for the three models. The first approach is confirmed as the square root of the AVE of each construct is higher than the correlations with other constructs. Additionally, all results of the HTMT test were below the threshold of highlighted in bold. This shows that the discriminant validity is not compromised because each variable is unique from the others.
Furthermore, as the answers were obtained from the same source, Common Method Variance (CMV) was examined via two tests
Table 1
Demographics characteristics of participants.
Demographics Categories Students Lecturers
N % N %
Gender Male 80 47.6 69 53.1
Female 88 52.4 61 46.9
Age 18-30 116 69.1 7 5.4
31-40 32 19 41 31.5
41-50 20 11.9 44 33.8
51-60 0 0 24 18.5
61+ 0 0 14 10.8
Educational High school or lower 46 27.4 5 3.9
level Bachelor 50 29.7 2 1.5
Master 43 25.6 22 16.9
Doctorate 26 15.5 98 75.4
Others 3 1.8 3 2.3
Field of study/ research Social Sciences (e.g. Business, Economics, Tourism and Hospitality, Psychology, Law, etc.) 134 79.8 110 84.6
Technology (e.g. Engineering, Robotics, Computer Science, Mechanics, etc.) 15 8.9 12 9.2
Arts & Humanities (e.g. Architecture, History, Literature, Music, Philosophy, etc.) 12 7.1 4 3.1
Life Sciences & Biomedicine (e.g. Biology, Medicine, Agriculture, etc.) 5 3 2 1.5
Physical Sciences (e.g. Astronomy, Chemistry, Physics, Mathematics, etc.) 2 1.2 2 1.5
Participants country Bulgaria 26 15.5 6 4.6
Indonesia 20 11.9 8 6.2
Portugal 13 7.7 10 7.7
United States 10 6.0 13 10.0
Finland 15 8.9 5 3.8
Oman 10 6.0 6 4.6
Poland 10 6.0 4 3.1
United Kingdom 4 2.4 9 6.9
India 5 3.0 5 3.8
Pakistan 4 2.4 4 3.1
Spain 0 0.0 8 6.2
Malaysia 5 3.0 2 1.5
Egypt 4 2.4 3 2.3
Turkey 2 1.2 5 3.8
Greece 2 1.2 4 3.1
Netherlands 2 1.2 4 3.1
Sweden 2 1.2 3 2.3
Taiwan 3 1.8 1 0.8
Australia 1 0.6 3 2.3
Others 30 17.9 27 20.8
Total 168 100.0 130 100.0
(Harman’s single-factor approach and full collinearity VIFs) and both were met confirming internal consistency as total variance explained by a single factor was less than 50% [55] and the VIF values were lower than 5, respectively [56]. Thus, the requirements for the measurement model have been met allowing the second step to be checked.

4.2. Structural models

The path coefficients of the models of the overall sample, lecturers and students are presented in Figs. 2-4, respectively. The strengths of GenAI and the benefits of GenAI had positive and significant impacts on all three factors of the TPB model (attitude, subjective norms, and perceived behavioural control) with -values below the threshold of 0.05 in all three models (see Table 6). Therefore, H1a, H1b, H1c, H2a, H 2 b , and H 2 c are supported in all three models.
The results revealed that the impact of the weaknesses of GenAI on
attitude remained insignificant in all three models as the -values were above the threshold of 0.05 as shown in Figs. 2-4: total sample ( ; ), lecturers ( ), and students ( 0.26). Similarly, the impact of the weaknesses of GenAI on subjective norms remained insignificant in all three models: total sample ( ; ), lecturers ( ), and students ( 0.44). In addition, its negative impact on perceived behavioural control was not supported in all three models: total sample ( ), lecturers ( ), students ( ) (Figs. 2 and 4). Therefore, H3a, H3b, and H3c were rejected.
Additionally, the findings indicated that the perceived risks of GenAI had a significant impact only on the attitudes in the overall model ( ) and lecturers’ model ( ); however, there was no significant relationship in the students’ model ( ; p ). H4a was thus accepted for both the lecturers’ sample and the total sample. In addition, the results did not support the significant and negative relationship between perceived risks of GenAI and the subjective norms in the three models: total sample ( ), lecturers ( ), and students ( ). Thus, H4b was rejected. The impact of perceived risks of GenAI on perceived behavioural control was significant in the overall model ( ; 0.04 ) but insignificant for lecturers ( ) and students ( ). Therefore, H4c was accepted for the whole sample only.
The empirical results also indicated that the associations between the three core TPB variables (i.e., attitude, subjective norms, and perceived behavioural control) and the intention to use GenAI tools were significant and positive in all three models. Thus, H5, H6, and H7 were all accepted. Finally, there was a significant and positive link between the intention to use and actual use of GenAI in all three models. Thus, H8 was supported (see Table 6).

5. Discussion and conclusions

5.1. Discussion and theoretical implications

Underpinned by the TPB, the present research examined the nexus between perceived strengths, advantages, weaknesses, and risks of GenAI tools and the core variables of the TPB model, namely attitude, subjective norms, and perceived behavioural control. This paper also investigated how the fundamental variables of the TPB model could impact the intention toward adopting GenAI tools, and how the latter could influence the actual usage of GenAI tools by both lecturers and students in higher education institutions (HEIs). Overall, lecturers’ and students’ attitudes, subjective norms, and perceived behavioural control were significantly and positively affected by the perceived strengths and benefits of GenAI tools; however, lecturers’ and students’ perceptions of the weaknesses and risks of these tools vary.
The empirical results indicated that the strengths and benefits of GenAI applications had a positive and significant impact on all three of the TPB core components (i.e., attitude, subjective norms, and perceived behavioural control) in all three models (overall sample, lecturers, and students). These findings imply that the more attainable strengths and benefits of GenAI technologies used in HEIs, the greater the positive impacts they have on lecturers’ and students’ attitudes, subjective norms, and perceived behavioural control. These results underscore the importance of showcasing the advantages and attributes of GenAI apps to promote positive attitudes, subjective norms, and user perceptions of their usability-all of which can eventually result in the acceptance and effective use of these tools in HEIs. These results support the findings of prior studies articulating that perceived strengths and benefits of technologies could positively shape educators’ attitudes (e.g., Ref. [24]), subjective norms (e.g., Ref. [26]), and perceived control (e.g., Ref. [22]) toward the adoption of such tools.
The empirical findings did not support some of the hypotheses related to the connections between the weaknesses of GenAI and the
Table 2
Descriptive statistics and assessment results of the measurement models.
Total sample ( ) Lecturers ( ) Students ( )
Construct/items Indicator loading Composite reliability AVE Indicator loading Composite reliability AVE Indicator loading Composite reliability AVE
Strengths of Gen-AI 0.876 0.640 0.811 0.799 0.883 0.655
STRN1 0.768 0.794 0.752
STRN2 0.850 0.804 0.840
STRN3 0.826 0.794 0.879
STRN4 0.752 0.804 0.760
Benefits of Gen-AI 0.938 0.791 0.915 0.893 0.938 0.790
BNFT1 0.892 0.916 0.878
BNFT2 0.863 0.839 0.879
BNFT3 0.889 0.902 0.881
BNFT4 0.913 0.915 0.916
Weaknesses of Gen-AI 0.873 0.580 0.866 0.776 0.854 0.593
WEAK1 0.816 0.795 0.782
WEAK2 0.822 0.877 0.766
WEAK3 0.719 0.780 0.770
WEAK4 0.740 0.708 0.762
WEAK5 0.702 0.745 NA
WEAK6 NA 0.739 NA
Risks of Gen-AI 0.918 0.848 0.858 0.936 0.863 0.678
RSK1 0.921 0.936 0.863
RSK2 0.921 0.936 0.857
RSK3 NA NA 0.745
Attitude 0.932 0.697 0.931 0.865 0.933 0.665
ATTD1 0.787 0.747 0.830
ATTD2 0.843 0.856 0.818
ATTD3 0.845 0.882 0.785
ATTD4 0.848 0.924 0.804
ATTD5 0.859 0.872 0.872
ATTD6 0.825 0.896 0.769
ATTD7 NA NA 0.827
Subjective norms 0.930 0.816 0.921 0.929 0.909 0.770
SUBJ1 0.906 0.930 0.882
SUBJ2 0.912 0.928 0.897
SUBJ3 0.891 0.930 0.851
Perceived behavioural control 0.893 0.736 0.827 0.862 0.892 0.733
PRCV1 0.827 0.817 0.836
PRCV2 0.867 0.884 0.854
PRCV3 0.879 0.884 0.877
Intention to use Gen-AI 0.898 0.747 0.857 0.882 0.887 0.724
INTN1 0.856 0.876 0.843
INTN2 0.872 0.884 0.862
INTN3 0.864 0.886 0.848
Actual use of Gen-AI 0.922 0.855 0.858 0.936 0.909 0.833
ACTU1 0.924 0.936 0.913
ACTU2 0.924 0.936 0.913
three variables of the TPB. Although previous studies have emphasised GenAI’s weaknesses and opportunities for abuse (e.g., Ref. [15,16]), this study found that they were not crucial in determining lecturers’ and students’ attitudes, social norms, or perceptions of behavioural control but the strengths and benefits of GenAI were. Additionally, the analysis produced mixed results regarding the impact of perceived risks of GenAI on attitude, subjective norms, and perceived behavioural control (only 3 out of 9 hypotheses related to perceived risk were supported in the three models). It implies that lecturers and students differ in attitudes, perceptions of behavioural control, and adherence to social norms when it comes to how they understand and react to risks around GenAI applications (such as ChatGPT). In this sense, the results provide only partial support to prior studies (e.g., Ref. [4,9,16,37]) that have highlighted the perceived risks related to the adoption of GenAI applications, including ChatGPT. Although the existing literature (e.g., Ref. [16]) has outlined several difficulties and ethical concerns surrounding the application of GenAI in learning environments, such as data security, bias, authenticity, and pedagogical coherence, this study showed that these risks are not very important to the lecturers and students. In fact, the perceived strengths and benefits of using GenAI in teaching and research are much more important drivers of GenAI adoption in educational setting than their weaknesses and risks. The novelty of these tools, the potential
advantages they give to the students and lecturers in terms of time savings and productivity, and the significant improvements in the quality of GenAI’s outputs over time might be the reasons why the strengths and benefits of GenAI have greater importance for the respondents than the weaknesses and risks associated with these tools but future research needs to provide a definitive answer to this question.
The findings also depicted that the three core TPB variables (i.e., attitude, subjective norms, and perceived behavioural control) had a positive and significant link with the intention to use GenAI tools by both lecturers and students. To encourage the use and integration of GenAI tools in higher education, it is critical to cultivate positive attitudes, subjective norms, and perceived behavioural control toward these tools among both lecturers and students. The intentions of both lecturers and students to utilise these tools may be increased by highlighting the advantages, benefits, and efficacy of these tools. This will improve students’ and lecturers’ attitudes, subjective norms, and perceived control. These results are consistent with the principles of the TPB, which holds that a person’s intent to engage in a certain behaviour-in this case, utilising GenAI tools-is greatly influenced by these three variables [20]. The results also showed that there is a strong correlation between the intention to employ GenAI technologies and their actual application. This finding aligns with the general understanding in behavioural
Table 3
Discriminant validity (Total sample).
Construct Fornell and Larcker [53]
STRN BNFT WEAK RSK ATTD SUBJ PRCV INIT ACTU
STRN (0.800)*
BNFT 0.657 (0.889)
WEAK -0.079 -0.017 (0.762)
RSK -0.058 -0.062 0.334 (0.921)
ATTD 0.554 0.575 -0.056 -0.149 (0.835)
SUBJ 0.335 0.458 0.007 -0.039 0.420 (0.903)
PRCV 0.567 0.591 0.084 -0.032 0.503 0.379 (0.858)
INTN 0.586 0.668 -0.009 -0.095 0.582 0.527 0.709 (0.864)
ACTU 0.398 0.491 0.012 -0.111 0.472 0.506 0.540 0.684 (0.924)
HTMT ratios STRN BNFT WEAK RSK ATTD SUBJ PRCV INIT ACTU
STRN
BNFT 0.766**
WEAK 0.168 0.102
RSK 0.075 0.078 0.408
ATTD 0.645 0.631 0.144 0.172
SUBJ 0.394 0.509 0.066 0.053 0.468
PRCV 0.698 0.680 0.166 0.088 0.580 0.442
INTN 0.717 0.767 0.139 0.115 0.670 0.615 0.858
ACTU 0.487 0.564 0.056 0.135 0.543 0.590 0.651 0.823
*Numbers in brackets reflect the square root of average values (AVEs), whereas the other numbers indicate the correlations among factors.
** Bold values HTMT ratio that are lower than 0.90 indicate that: that variable is distinct from other variables confirming its uniqueness.
Table 4
Discriminant validity (Lecturers).
Construct Fornell and Larcker [53]
STRN BNFT WEAK RSK ATTD SUBJ PRCV INIT ACTU
STRN (0.799)
BNFT 0.549 (0.893)
WEAK -0.112 -0.009 (0.776)
RSK -0.167 -0.187 0.282 (0.936)
ATTD 0.596 0.536 -0.062 -0.360 (0.865)
SUBJ 0.451 0.535 0.051 -0.154 0.416 (0.862)
PRCV 0.516 0.657 -0.033 -0.244 0.602 0.638 (0.882)
INTN 0.460 0.586 -0.013 -0.251 0.526 0.589 0.806 (0.936)
ACTU 0.297 0.452 0.008 -0.174 0.426 0.303 0.488 0.527 (0.814)
HTMT ratios BNFT WEAK RSK ATTD SUBJ PRCV INIT ACTU
STRN
BNFT 0.638
WEAK 0.159 0.088
RSK 0.200 0.211 0.330
ATTD 0.686 0.582 0.127 0.405
SUBJ 0.549 0.610 0.193 0.180 0.472
PRCV 0.618 0.741 0.093 0.285 0.676 0.754
INTN 0.551 0.661 0.094 0.293 0.590 0.695 0.940
ACTU 0.360 0.519 0.098 0.336 0.495 0.353 0.573 0.618
*Numbers in brackets reflect the square root of average values (AVEs), whereas the other numbers indicate the correlations among factors.
** Bold values HTMT ratio that are lower than 0.90 indicate that: that variable is distinct from other variables confirming its uniqueness.
psychology that intentions often serve as a reliable determinant of subsequent behaviour. It is a valuable insight for understanding the factors influencing the adoption of GenAI tools in practice by both lecturers and students (e.g., Ref. [57,58]).
Theoretically, this work complements earlier research that emphasised the key concerns surrounding the application of GenAI tools in research and education [9,37]. Previous studies in this area focused on the main variables influencing the use of these instruments in research and education within various settings (e.g., Ref. [57-59]). However, the current study is one of the first attempts to incorporate several crucial elements, such as the strengths, benefits, weaknesses, and risks of using GenAI tools within a structural model to measure the most significant factors influencing the actual utilisation of these tools based on an international sample of lecturers/researchers and students. Furthermore, this study enhances the theory by extending the TPB model to explicitly demonstrate the main drivers for deploying GenAI applications such as
ChatGPT in HEIs. Additionally, this is one of the first studies to compare the views of lecturers and students on the critical elements impacting the use of GenAI tools.

5.2. Managerial and policy implications

The study presents a set of practical implications for concerned stakeholders at HEIs (e.g., students, lecturers and HEI administrators). The empirical findings reveal that improving lecturers’ and students’ perspectives of the advantages and benefits of implementing GenAI tools could be associated with a more positive attitude, subjective norms, and perceived behavioural control toward the use of such tools in research and education at HEIs. This could be accomplished by developing and deploying efficient procedures and mechanisms to spread knowledge and awareness about the use of GenAI tools in teaching and research. In addition, HEIs need to organise workshops and awareness-building
Table 5
Discriminant validity (Students).
Construct Fornell and Larcker [53]
STRN BNFT WEAK RSK ATTD SUBJ PRCV INIT ACTU
STRN (0.712)
BNFT 0.688 (0.889)
WEAK -0.077 0.002 (0.692)
RSK 0.054 0.077 0.489 (0.663)
ATTD 0.537 0.601 0.004 0.084 (0.816)
SUBJ 0.417 0.457 0.038 0.217 0.388 (0.877)
PRCV 0.605 0.679 0.021 0.111 0.562 0.551 (0.851)
INTN 0.349 0.426 0.044 0.084 0.414 0.457 0.576 (0.913)
ACTU 0.620 0.629 0.115 0.148 0.582 0.448 0.766 0.499 (0.856)
HTMT ratios STRN BNFT WEAK RSK ATTD SUBJ PRCV INIT ACTU
STRN
BNFT 0.745
WEAK 0.183 0.106
RSK 0.066 0.120 0.569
ATTD 0.560 0.657 0.148 0.096
SUBJ 0.493 0.520 0.158 0.180 0.441
PRCV 0.693 0.791 0.125 0.097 0.652 0.666
INTN 0.341 0.500 0.084 0.113 0.484 0.554 0.715
ACTU 0.691 0.728 0.139 0.137 0.672 0.538 0.941 0.615
*Numbers in brackets reflect the square root of average values (AVEs), whereas the other numbers indicate the correlations among factors.
** Bold values HTMT ratio that are lower than 0.90 indicate that: that variable is distinct from other variables confirming its uniqueness.
Fig. 2. Structural Model (overall sample).
Fig. 3. Structural model (lecturers).
Fig. 4. Structural model (students).
Table 6
Summary of the hypotheses results.
Hypothesis Result
Overall sample Lecturers Students
H1a Perceived strengths of generative AI have a positive effect on attitude towards using generative AI. Accepted Accepted Accepted
H1b Perceived strengths of generative AI have a positive effect on subjective norms. Accepted Accepted Accepted
H1c Perceived strengths of generative AI have a positive effect on perceived behavioural control. Accepted Accepted Accepted
H2a Perceived benefits of generative AI have a positive effect on attitude towards using generative AI. Accepted Accepted Accepted
H2b Perceived benefits of generative AI have a positive effect on subjective norms. Accepted Accepted Accepted
H2c Perceived benefits of generative AI have a positive effect on perceived behavioural control. Accepted Accepted Accepted
H3a Perceived weaknesses of generative AI have a negative effect on attitude towards using generative AI. Rejected Rejected Rejected
H3b Perceived weaknesses of generative AI have a negative effect on subjective norms. Rejected Rejected Rejected
H3c Perceived weaknesses of generative AI have a negative effect on perceived behavioural control. Rejected Rejected Rejected
H4a Perceived risks of generative AI have a negative effect on attitude towards using generative AI. Accepted Accepted Rejected
H4b Perceived risks of generative AI have a negative effect on subjective norms. Rejected Rejected Rejected
H4c Perceived risks of generative AI have a negative effect on perceived behavioural control. Accepted Rejected Rejected
H5 Attitude towards using generative AI has a positive effect on the intention to use generative AI Accepted Accepted Accepted
H6 Subjective norms regarding generative AI have a positive effect on the intention to use generative AI Accepted Accepted Accepted
H7 Perceived control regarding generative AI has a positive effect on the intention to use generative AI Accepted Accepted Accepted
H8 Intention to use generative AI has a positive effect on actual use of generative AI Accepted Accepted Accepted
initiatives and training sessions that demonstrate the strengths and benefits of adopting GenAI applications, including ChatGPT, for teaching and research purposes.
The findings highlight the role of attitude, subjective norms, and perceived behavioural control in shaping the adoption of GenAI tools in education and research. Positive attitudes about GenAI tools among academics and students increase the likelihood that they will use such tools, underscoring the significance of promoting positive attitudes through educational programmes and emphasising the advantages of such tools in both education and research. Moreover, subjective norms that are shaped by academic communities’ and peers’ opinions can affect how socially accepted GenAI tools are. HEIs ought to cultivate a cooperative atmosphere that promotes the exchange of knowledge and highlights the combined benefits of employing GenAI tools in research and teaching environments. Furthermore, perceived behavioural control includes aspects such as perceived ease of use and technical expertise. To improve users’ confidence in using GenAI tools efficiently, HEIs must develop ethical standards for the responsible use of GenAI and provide relevant training programmes. HEIs need to establish a culture of acceptance and proficiency among lecturers and students by establishing an atmosphere that is favourable to the effective, efficient and ethical integration of GenAI tools in teaching and research. HEIs need to develop initiatives to encourage lecturers and students to use GenAI responsibly. Policymakers can help by developing legal regulations to
manage the application of GenAI in research and education to mitigate its negative impacts [41] by involving HEIs in the process [60]. Furthermore, to support inclusivity and diversity in research and education public authorities and HEIs need to ensure that GenAI technologies are available to all lecturers and students to ensure their competitiveness and employability, e.g. through institutional accounts to GenAI applications.

5.3. Limitations and future research directions

Although the current study provides several theoretical and practical contributions, some limitations offer valuable directions for future research. First, the current research relied on a quantitative approach using an online questionnaire to collect primary data from the targeted respondents (i.e., students and lecturers) from several countries. Future research could adopt a qualitative approach by using interviews or a mixed-method approach. Second, the current study model builds upon the TPB and expands it by including a group of variables (i.e., strengths of GenAI, benefits of GenAI, weaknesses of GenAI, and risks of GenAI). Future studies could develop a more comprehensive model by incorporating other theories such as the Diffusion of Innovation Theory [61], the technology acceptance model [62], or the unified theory of acceptance and use of technology [63]. Additional variables, such as stress, trust, and self-efficacy, can also be incorporated into the existing model.
Third, this study focused on higher education. Future research could focus on the use of GenAI in secondary education to yield insights into the variables influencing the perceptions and behaviours of high school students and teachers towards the applications of GenAI tools. Additionally, since the use of GenAI tools is the study’s outcome variable, further research can examine the effects of this use in a variety of areas, including levels of creative thinking, academic achievement, scientific productivity, and research ethics. Furthermore, future research may delve into the nuances of GenAI use in different countries, allowing for a more comprehensive understanding of potential linkages and distinct patterns across diverse cultural and socioeconomic contexts.

Declaration of generative AI in scientific writing

During the preparation of this work the authors used ChatGPT and Grammarly for proofreading and improve the readability of the text. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Declaration of interest statement

None.

CRediT authorship contribution statement

Stanislav Ivanov: Conceptualization, Methodology, Investigation, Visualization, Supervision, Writing – original draft, Writing – review & editing. Mohammad Soliman: Conceptualization, Methodology, Investigation, Writing – original draft, Writing – review & editing. Aarni Tuomi: Conceptualization, Methodology, Investigation, Writing original draft, Writing – review & editing. Nasser Alhamar Alkathiri: Conceptualization, Methodology, Investigation, Formal analysis, Visualization, Writing – original draft, Writing – review & editing. Alamir N. Al-Alawi: Conceptualization, Methodology, Investigation, Writing original draft, Writing – review & editing.

Data availability

Data will be made available on request.

Appendix 1. Measures

Variables Items Source
Strengths of generative AI
For the questions below think about the generative AI you mostly used:
I think the generative AI application:
STRN 1*** understood the nuances of human language
STRN interacted in a conversational and human-like way.
STRN could answer follow-up questions,
STRN 4*** … could admit its mistakes.
STRN 5 *** … could reject inappropriate requests.
STRN could keep track of the ongoing conversation.
STRN 7 … generated content that was useful to me
Developed by authors
Benefits of generative AI
For the questions below think about the generative AI you mostly used:
I think the generative AI application:
BNFT1 … enhanced the efficiency of my work
BNFT2 … improved the quality of work I do.
BNFT3 … helped me accomplish my tasks faster.
BNFT4 Overall, I find using the generative AI application to be advantageous in my work.
Hsu et al. [45] and expanded by the authors
Weaknesses of generative AI
For the questions below think about the generative AI you mostly used:
I think the generative AI application:
WEAK1 … was generating false information.
WEAK2*** … was not capable of ethical reasoning.
WEAK3 … lacked reliability about factual knowledge.
WEAK4* … struggled with providing proper referencing to the sources it was using.
WEAK5** … was not able to ask clarifying questions when given ambiguous prompts.
WEAK6 … was not delivering an adequate answer to my questions.
WEAK7 … struggled with generating responses to complex or abstract questions
Developed by authors
Risks of generative AI
For the questions below think about the generative AI you mostly used:
I think the generative AI application:
RSK1*** … generated responses that may have been biased.
RSK2*** … was using sensitive data I shared with it as training data.
RSK3*** … might replace many research-based jobs.
RSK4 … might decrease the credibility of my work
RSK5 … might decrease other people’s trust in my work
RSK6** … might be banned by my institution
RSK7*** … might be banned by academic journals
Developed by authors
Attitude
ATTD1- For me, using generative AI is extremely bad
ATTD2- For me, using generative AI is extremely undesirable
ATTD3- For me, using generative AI is extremely unpleasant
ATTD4- For me, using generative AI is extremely foolish
ATTD5- For me, using generative AI is extremely unfavourable
ATTD6- For me, using generative AI is extremely unenjoyable
ATTD7**- For me, using generative AI is extremely negative
[44]
Variables Items Source
Perceived behavioural control PRCV1- Whether or not I use generative AI while doing my work is completely up to me PRCV2- I am confident that if I want, I can use generative AI while doing my work [44]
PRCV3- I have resources, time, and opportunities to use generative AI while doing my work
Intention to use generative AI INTN1- It is worth it to use generative AI while doing my work [45]
INTN2- I will frequently use generative AI while doing my work in the future.
Actual use of generative AI ACTU1- I use generative AI on a daily basis [43]
ACTU2- I use generative AI frequently
*** Removed from all models due to low indicator loading.
** Removed from two models (total sample and students) due to low indicator loading.
*Removed only students model due to low indicator loading.

Appendix 2. Model fit and quality indices

Metric Overall Lectures Students Recommended value
Average path coefficient (APC) 0.251, P < 0.001 0.266, P < 0.001 0.255, P < 0.001
Average R-squared (ARS) 0.433, 0.474, P < 0.001 0.440, P < 0.001
Average adjusted R-squared (AARS) 0.426, P < 0.001
Average block VIF (AVIF) 1.372 1.273 1.508 ideally
Average full collinearity VIF (AFVIF) 1.989 2.104 2.141 ideally
Tenenhaus GoF (GoF) 0.568 0.595 0.560 Large
Simpson’s paradox ratio (SPR) 0.938 0.938 1.000 Acceptable if , Ideally
R-squared contribution ratio (RSCR) 0.998 0.996 1.000 Acceptable if , Ideally
Statistical suppression ratio (SSR) 1.000 0.938 0.938 Acceptable if
Nonlinear bivariate causality direction ratio (NLBCDR) 0.875 0.844 0.906 Acceptable if

References

[1] J. Wirtz, V. Pitardi, How intelligent automation, service robots, and AI will reshape service products and their delivery, Ital. J. Market. 2023 (2023) 289-300, https:// doi.org/10.1007/s43039-023-00076-1.
[2] O. Ali, P. Murray, M. Momin, F.S. Al-Anzi, The knowledge and innovation challenges of ChatGPT: a scoping review, Technol. Soc. 75 (2023) 102402, https:// doi.org/10.1016/j.techsoc.2023.102402.
[3] S.A. Bin-Nashwan, M. Sadallah, M. Bouteraa, Use of ChatGPT in academia: academic integrity hangs in the balance, Technol. Soc. 75 (2023) 102370, https:// doi.org/10.1016/j.techsoc.2023.102370.
[4] Y.K. Dwivedi, N. Kshetri, L. Hughes, E.L. Slade, A. Jeyaraj, A.K. Kar, A. M. Baabdullah, A. Koohang, V. Raghavan, M. Ahuja, “So what if ChatGPT wrote it?” Multidisciplinary perspectives on opportunities, challenges and implications of generative conversational AI for research, practice and policy, Int. J. Inf. Manag. 71 (2023) 102642, https://doi.org/10.1016/j.ijinfomgt.2023.102642.
[5] S. Rice, S.R. Crouse, S.R. Winter, C. Rice, The advantages and limitations of using ChatGPT to enhance technological research, Technol. Soc. 76 (2024) 102426, https://doi.org/10.1016/j.techsoc.2023.102426.
[6] H.S. Sætra, Generative AI: here to stay, but for good? Technol. Soc. 75 (2023) 102372 https://doi.org/10.1016/j.techsoc.2023.102372.
[7] A. Susarla, R. Gopal, J.B. Thatcher, S. Sarker, The Janus effect of generative AI: charting the path for responsible conduct of scholarly activities in information systems, Inf. Syst. Res. 34 (2) (2023) 399-408, https://doi.org/10.1287/isre.2023. ed.v34.n2.
[8] R. Vogler, 2023 – an AI university space Odyssey, Retrieved 15th February 2024 from, ROBONOMICS: J. Autom. Econ. 5 (2024) 55, https://journal.robonomics. science/index.php/rj/article/view/55.
[9] S. Ivanov, M. Soliman, Game of algorithms: ChatGPT implications for the future of tourism education and research, J. Tourism Futur. 9 (2) (2023) 214-221, https:// doi.org/10.1108/JTF-02-2023-0038.
[10] T.K. Chiu, The impact of Generative AI (GenAI) on practices, policies and research direction in education: a case of ChatGPT and Midjourney, Interact. Learn. Environ. (2023) 1-17, https://doi.org/10.1080/10494820.2023.2253861.
[11] M. Farrokhnia, S.K. Banihashem, O. Noroozi, A. Wals, A SWOT analysis of ChatGPT: implications for educational practice and research, Innovat. Educ. Teach. Int. (2023) 1-15, https://doi.org/10.1080/14703297.2023.2195846.
[12] A. Gilson, C. Safranek, T. Huang, V. Socrates, L. Chi, R.A. Taylor, D. Chartash, How well does ChatGPT do when taking the medical licensing exams? The implications of large language models for medical education and knowledge assessment, medRxiv (2022) 1-9, https://doi.org/10.1101/2022.12.23.22283901.
[13] F.M. Megahed, Y.-J. Chen, J.A. Ferris, S. Knoth, L.A. Jones-Farmer, How generative AI models such as ChatGPT can be (mis)used in SPC practice, education, and research? An exploratory study, Qual. Eng. 36 (2) (2024) 287-315, https://doi. org/10.1080/08982112.2023.2206479.
[14] S. Biswas, ChatGPT and the future of medical writing, Radiology 307 (2) (2023) e223312, https://doi.org/10.1148/radiol.223312.
[15] D.R. Cotton, P.A. Cotton, J.R. Shipway, Chatting and cheating: ensuring academic integrity in the era of ChatGPT, Innovat. Educ. Teach. Int. 61 (2) (2024) 228-239, https://doi.org/10.1080/14703297.2023.2190148.
[16] E. Kasneci, K. Seßler, S. Küchemann, M. Bannert, D. Dementieva, F. Fischer, U. Gasser, G. Groh, S. Günnemann, E. Hüllermeier, ChatGPT for good? On opportunities and challenges of large language models for education, Learn. Indiv Differ 103 (2023) 102274, https://doi.org/10.1016/j.lindif.2023.102274.
[17] A. Strzelecki, S. ElArabawy, Investigation of the moderation effect of gender and study level on the acceptance and use of generative AI by higher education students: comparative evidence from Poland and Egypt, Br. J. Educ. Technol. (2024), https://doi.org/10.1111/bjet. 13425.
[18] M. Jaboob, M. Hazaimeh, A.M. Al-Ansi, Integration of generative AI techniques and applications in student behavior and cognitive achievement in Arab higher education, Int. J. Hum. Comput. Interact. (2024) 1-14, https://doi.org/10.1080/ 10447318.2023.2300016.
[19] Y. Wang, W. Zhang, Factors influencing the adoption of generative AI for art designing among Chinese generation Z: a structural equation modeling approach, IEEE Access 11 (2023) 143272-143284, https://doi.org/10.1109/ ACCESS. 2023.3342055.
[20] I. Ajzen, The theory of planned behavior, Organ. Behav. Hum. Decis. Process. 50 (2) (1991) 179-211, https://doi.org/10.1016/0749-5978(91)90020-T.
[21] I. Ajzen, The theory of planned behavior: frequently asked questions, Hum. Behav. Emerg. Technol. 2 (4) (2020) 314-324, https://doi.org/10.1002/hbe2.195.
[22] H. Knauder, C. Koschmieder, Individualized student support in primary school teaching: a review of influencing factors using the Theory of Planned Behavior (TPB), Teach. Teach. Educ. 77 (2019) 66-76, https://doi.org/10.1016/j. tate.2018.09.012.
[23] T. Teo, C. Beng Lee, Explaining the intention to use technology among student teachers: an application of the Theory of Planned Behavior (TPB), Campus-Wide Inf. Syst. 27 (2) (2010) 60-67, https://doi.org/10.1108/10650741011033035.
[24] M. Bosnjak, I. Ajzen, P. Schmidt, The theory of planned behavior: selected recent advances and applications, Eur. J. Psychol. 16 (3) (2020) 352-356, https://doi. org/10.5964/ejop.v16i3.3107.
[25] M. Conner, C.J. Armitage, Extending the theory of planned behavior: a review and avenues for further research, J. Appl. Soc. Psychol. 28 (15) (1998) 1429-1464, https://doi.org/10.1111/j.1559-1816.1998.tb01685.x.
[26] Y. Wang, C. Dong, X. Zhang, Improving MOOC learning performance in China: an analysis of factors from the TAM and TPB, Comput. Appl. Eng. Educ. 28 (6) (2020) 1421-1433, https://doi.org/10.1002/cae.22310.
[27] J. Cheon, S. Lee, S.M. Crooks, J. Song, An investigation of mobile learning readiness in higher education based on the theory of planned behavior, Comput. Educ. 59 (3) (2012) 1054-1064, https://doi.org/10.1016/j.compedu.2012.04.015.
[28] K.M. White, I. Thomas, K.L. Johnston, M.K. Hyde, Predicting attendance at peerassisted study sessions for statistics: role identity and the theory of planned behavior, J. Soc. Psychol. 148 (4) (2008) 473-492, https://doi.org/10.3200/ SOCP.148.4.473-492.
[29] D. Baidoo-Anu, L.O. Ansah, Education in the era of generative artificial intelligence (AI): understanding the potential benefits of ChatGPT in promoting teaching and learning, J. AI 7 (1) (2023) 52-62, https://doi.org/10.61969/jai.1337500.
[30] J. Qadir, Engineering education in the era of ChatGPT: promise and pitfalls of generative AI for education, in: 2023 IEEE Global Engineering Education Conference (EDUCON), Kuwait, Kuwait, 2023, pp. 1-9, https://doi.org/10.1109/ EDUCON54358.2023.10125121.
[31] Stanford, Responsible AI at Stanford, 2023. Retrieved 16th December 2023 from, https://uit.stanford.edu/security/responsibleai.
[32] UNESCO, Generative Artificial Intelligence in Education: what Are the Opportunities and Challenges, 2023. Retrieved 16th December 2023 from, https:// www.unesco.org/en/articles/generative-artificial-intelligence-education-what-are -opportunities-and-challenges.
[33] J. Walmsley, Artificial intelligence and the value of transparency, AI Soc. 36 (2) (2021) 585-595, https://doi.org/10.1007/s00146-020-01066-z.
[34] S. Noy, W. Zhang, Experimental evidence on the productivity effects of generative artificial intelligence, SSRN 4375283, https://doi.org/10.2139/ssrn.4375283, 2023.
[35] M. Shanahan, K. McDonell, L. Reynolds, Role play with large language models, Nature 623 (2023) 493-498, https://doi.org/10.1038/s41586-023-06647-8.
[36] L. Kohnke, B.L. Moorhouse, D. Zou, ChatGPT for language teaching and learning, 00336882231162868, RELC J. (2023).
[37] I. Carvalho, S. Ivanov, ChatGPT for tourism: applications, benefits and risks, Tour. Rev. 79 (2) (2024) 290-303, https://doi.org/10.1108/TR-02-2023-0088.
[38] E. Bender, T. Gebru, A. McMillan-Major, S. Shmitchell, On the dangers of stochastic parrots: can language models be too big?, in: FAccT ’21: Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, 2021, pp. 610-623, https://doi.org/10.1145/3442188.3445922.
[39] J. Baker-Brunnbauer, TAII framework for trustworthy AI systems, Retrieved 23rd June 2023 from, ROBONOMICS: J. Autom. Econ. 2 (2021) 17, https://journal. robonomics.science/index.php/rj/article/view/17.
[40] Z. Li, The dark side of chatgpt: legal and ethical challenges from stochastic parrots and hallucination, arXiv preprint arXiv:2304.14347, https://doi.org/10.4855 0/arXiv.2304.14347, 2023.
[41] S. Ivanov, The dark side of artificial intelligence in higher education, Serv. Ind. J. 43 (15-16) (2023) 1055-1082, https://doi.org/10.1080/ 02642069.2023 .2258799.
[42] J.F. Hair, C.M. Ringle, M. Sarstedt, PLS-SEM: indeed a silver bullet, J. Market. Theor. Pract. 19 (2) (2011) 139-152, https://doi.org/10.2753/MTP10696679190202.
[43] W.H. DeLone, E.R. McLean, The DeLone and McLean model of information systems success: a ten-year update, J. Manag. Inf. Syst. 19 (4) (2003) 9-30, https://doi.org/ 10.1080/07421222.2003.11045748.
[44] H. Han, L.-T.J. Hsu, C. Sheu, Application of the theory of planned behavior to green hotel choice: testing the effect of environmental friendly activities, Tourism Manag. 31 (3) (2010) 325-334, https://doi.org/10.1016/j.tourman.2009.03.013.
[45] C.-L. Hsu, H.-P. Lu, H.-H. Hsu, Adoption of the mobile Internet: an empirical study of multimedia message service (MMS), Omega 35 (6) (2007) 715-726, https://doi. org/10.1016/j.omega.2006.03.005.
[46] N. Kock, WarpPLS User Manual: Version 8.0 Scriptwarp Systems: Laredo TX, USA, 2022. Retrieved 16th December 2023 from, https://scriptwarp.com/warppls/U serManual_v_8_0.pdf.
[47] N. Kock, L. Gaskins, The mediating role of voice and accountability in the relationship between Internet diffusion and government corruption in Latin America and Sub-Saharan Africa, Inf. Technol. Dev. 20 (1) (2014) 23-43, https:// doi.org/10.1080/02681102.2013.832129.
[48] J. Henseler, G. Hubona, P.A. Ray, Using PLS path modeling in new technology research: updated guidelines, Ind. Manag. Data Syst. 116 (1) (2016) 2-20, https:// doi.org/10.1108/IMDS-09-2015-0382.
[49] M. Soliman, S. Ivanov, C. Webster, The psychological impacts of COVID-19 outbreak on research productivity: a comparative study of tourism and nontourism scholars, J. Tour. Dev. 35 (2021) 23-52, https://doi.org/10.34624/rtd. v0i35.24616.
[50] M. Soliman, R. Sinha, F. Di Virgilio, M.J. Sousa, R. Figueiredo, Emotional intelligence outcomes in higher education institutions: empirical evidence from a Western context, 00332941231197165, Psychol. Rep. (2023), https://doi.org/ 10.1177/00332941231197165.
[51] M.A.Q. Tran, T. Vo-Thanh, M. Soliman, B. Khoury, N.N.T. Chau, Self-compassion, mindfulness, stress, and self-esteem among Vietnamese university students: psychological well-being and positive emotion as mediators, Mindfulness 13 (10) (2022) 2574-2586, https://doi.org/10.1007/s12671-022-01980-x.
[52] J.F. Hair, M.C. Howard, C. Nitzl, Assessing measurement model quality in PLS-SEM using confirmatory composite analysis, J. Bus. Res. 109 (2020) 101-110, https:// doi.org/10.1016/j.jbusres.2019.11.069.
[53] C. Fornell, D.F. Larcker, Structural equation models with unobservable variables and measurement error: Algebra and statistics, J. Market. Res. 18 (3) (1981) 382-388, https://doi.org/10.1177/002224378101800313.
[54] J. Henseler, C.M. Ringle, M. Sarstedt, A new criterion for assessing discriminant validity in variance-based structural equation modeling, J. Acad. Market. Sci. 43 (2015) 115-135, https://doi.org/10.1007/s11747-014-0403-8.
[55] P.M. Podsakoff, S.B. MacKenzie, J.-Y. Lee, N.P. Podsakoff, Common method biases in behavioral research: a critical review of the literature and recommended remedies, J. Appl. Psychol. 88 (5) (2003) 879, https://doi.org/10.1037/00219010.88.5.879.
[56] F. Kock, A. Berbekova, A.G. Assaf, Understanding and managing the threat of common method bias: detection, prevention and control, Tourism Manag. 86 (2021) 104330, https://doi.org/10.1016/j.tourman.2021.104330.
[57] T. Gundu, Chatbots: a framework for improving information security behaviours using ChatGPT, in: S. Furnell, N. Clarke (Eds.), Human Aspects of Information Security and Assurance, IFIP Advances in Information and Communication Technology, vol. 674, Springer, Cham, 2023, pp. 418-431, https://doi.org/ 10.1007/978-3-031-38530-8_33.
[58] C.S. Shah, S. Mathur, S.K. Vishnoi, Continuance intention of ChatGPT use by students, TDIT 2023, in: S.K. Sharma, Y.K. Dwivedi, B. Metri, B. Lal, A. Elbanna (Eds.), Transfer, Diffusion and Adoption of Next-Generation Digital Technologies, IFIP Advances in Information and Communication Technology, Springer, Cham, 2024, pp. 159-175, https://doi.org/10.1007/978-3-031-50188-3_14.
[59] A.M. Al-Zahrani, The impact of generative AI tools on researchers and research: implications for academia in higher education, Innovat. Educ. Teach. Int. (2023) 1-15, https://doi.org/10.1080/14703297.2023.2271445.
[60] V. Dubljević, Colleges and universities are important stakeholders for regulating large language models and other emerging AI, Technol. Soc. 76 (2024) 102480, https://doi.org/10.1016/j.techsoc.2024.102480.
[61] E.M. Rogers, Diffusion of Innovations, 1983, third ed., The Free Press, London, 1962.
[62] V. Venkatesh, F.D. Davis, A theoretical extension of the technology acceptance model: four longitudinal field studies, Manag. Sci. 46 (2) (2000) 186-204, https:// doi.org/10.1287/mnsc.46.2.186.11926.
[63] V. Venkatesh, M.G. Morris, G.B. Davis, F.D. Davis, User acceptance of information technology: toward a unified view, MIS Q. (2003) 425-478, https://doi.org/ .
Dr. Stanislav Ivanov is currently Professor and Vice-Rector (Research) at Varna University of Management, Bulgaria (http://www.vum.bg) and Director of Zangador Research Institute (https://www.zangador.institute/en/). Prof. Ivanov is the Founder and Editor-in-chief of two academic journals: European Journal of Tourism Research (http://ejtr.vumk.eu) and ROBONOMICS: The Journal of the Automated Economy (https ://journal.robonomics.science). His research interests include robonomics, robots in tourism/hospitality, the economics of technology, etc. For more information about Prof. Ivanov please visit his personal website: http://www.stanislavivanov.com.
Dr. Mohammad Soliman is currently the Head of Research & Consultation Department at the University of Technology and Applied Sciences, Salalah, Oman. He is also a Full Professor at the Faculty of Tourism & Hotels, Fayoum University, Egypt. He has published multiple papers in high-rank journals indexed in WoS and Scopus. Additionally, he sits on the editorial board of different academic journals and serves as a reviewer for several top-tier journals. He has successfully supervised and examined several master’s and PhD theses. His research interests include tourism marketing, consumer behaviour, branding, AI-enabled education and marketing, tourism management, PLS-SEM, bibliometrics, and literature review studies.
Dr. Aarni Tuomi is Senior Lecturer at Haaga-Helia University of Applied Sciences. His research, teaching and consultancy projects explore the intersection of emerging technologies and service business. His research has explored e.g. service robotics, artificial intelligence, digital platforms and food technology, as well as experience design and innovation. His work is regularly featured in industry trade magazines and his research has been published in top academic journals, e.g. Annals of Tourism Research, Psychology & Marketing.
Dr. Nasser Alhamar Alkathiri is an Assistant Professor and the Deputy Assistant Vice Chancellor for Postgraduate Studies, Scientific Research and Innovation, University of Technology and Applied Sciences, Salalah, Oman. Dr. Nasser holds a PhD in Knowledge Transfer from Plymouth University, the UK. He also holds a Master degree in International Business from Sydney University, Australia. His research interests include knowledge management, knowledge transfer, international, business, entrepreneurship and staff localization. He has published multiple papers in reputed journals indexed in WoS and Scopus (e.g. Journal of Knowledge Management, International Journal of Finance & Economics and International Journal of Contemporary Hospitality Management). He has several participations in local and international conferences.
Dr. Al-Amir obtained his bachelor’s degree from the Sultanate of Oman and pursued his master’s and doctoral degrees in the United Kingdom. Over the course of 23 years, he held various positions. He currently holds the position of Assistant Vice Chancellor of the University of Technology and Applied Sciences in Ibri. Dr. Al-Amir’s area of interest is finance, entrepreneurship, tourism and artificial intelligence. He succeeded in working on two research projects funded by the Scientific Research Council of Oman.

    • Corresponding author. Varna University of Management, 13A Oborishte Str., 9000 Varna, Bulgaria.
    E-mail addresses: stanislav.ivanov@vumk.eu, info@zangador.institute (S. Ivanov), msoliman.sal@cas.edu.om (M. Soliman), Aarni.Tuomi@haaga-helia.fi (A. Tuomi), nasser2014.sal@cas.edu.om (N.A. Alkathiri), alameer.alalawi@utas.edu.om (A.N. Al-Alawi).
    web: http://stanislavivanov.com/.