<![CDATA[This qualitative study evaluates ChatGPT-4o's spoken Arabic capabilities and grammatical mapping accuracy across two distinct devices using a single foundational Arabic sentence prompt ("اللغة العربية هي لغة الجنة"). The evaluation criteria encompassed diacritical precision (harakat), phonetic articulation (makhraj), syntactic comprehension (nahw), and I'rab parsing. We cross-verified data generated through structured dialogues against Al-Nahw Al-Wadih and validated it with an Arabic linguistics expert. The empirical findings indicate that ChatGPT-4o achieved uniform excellence, scoring 4 out of 4 in diacritical rendering and phonetic articulation across both devices. However, in systematic syntactic parsing (I'rab), the model achieved measured grammatical accuracy rates of 78.4% on Device I using an Arabic interface and 74.5% on Device II using an English interface. These findings show that while AI can support Arabic conversational practice and basic linguistic inquiries, its structural parsing performance remains constrained by default system language configurations, hardware variations, and overall sentence complexity.]]>
Copyrights © 2027
