Arabic Language Proofreading Efficiency in Three AI Applications: A Comparative Critical Study

Objectives: This study aimed to evaluate the efficiency of three AI-based Proofreading applications by focusing on grammar, morphology, spelling, and context, along with additional “complementary” proofreading aspects. The study sought to highlight each application’s strengths to promote their use, identify areas of weakness for improvement, analyze the challenges they face, and propose solutions.
Methodology: The researcher employed a descriptive, analytical, applied, critical, and comparative approach. Three applications—Sakhr, Qalam, and Lisan—Were chosen based on their popularity, reliance on AI, and claimed proofreading accuracy. The applications were tested using carefully constructed sentences designed to cover grammatical, morphological, spelling, and contextual issues. Their performance was documented in tables that displayed success rates for each level, followed by quantitative and qualitative analysis.
Results: The findings showed a notable success rate in spelling correction, reaching up to 70% for some applications. However, context-related errors remained high across all three. Sahehly performed best in grammar, while Qalam outperformed others in spelling. These results suggest that the applications can be used together to improve overall proofreading quality. Nonetheless, handling context remains the most significant challenge for all three tools.
Conclusion: The study concluded that these AI-based proofreading applications can enhance specific aspects of proofreading, especially spelling and grammar. However, further development is needed to handle contextual features more effectively. Currently, these tools should be viewed as supportive aids to human proofreading rather than complete replacements.