In brief
Taibah University in Medina closed its 6th International Conference on Artificial Intelligence and its technologies, showcasing studies on Arabic language model reliability, a lip‑reading accuracy test, medical query triage, manuscript digitization, and legal compliance. The event highlighted collaboration with Malaysia’s national university and cemented the university’s role in advancing AI for Arabic contexts.
Taibah University in Medina finished its sixth International Conference on Artificial Intelligence and its technologies (MCAIT 2026) on November 6, 2026 after three days of talks and parallel research sessions.
The program explored AI applications in health, education, cybersecurity, and smart‑city planning, with special emphasis on tools that work with Arabic language and culture.
One key finding was a set of 17 studies led by researcher Reem Al Tamimi that examined how well Arabic language models grasp different dialects. The work showed large gaps in understanding and suggested new standards for measuring model trustworthiness.
A second highlight was a lip‑reading experiment that taught a deep‑learning system to recognize four Arabic words—“yes,” “no,” “sorry,” and “thanks”—based solely on visual mouth movements. The prototype reached 87.02 % accuracy and improved to 78.01 % when an extra class for unknown words was added.
In medical AI, researchers Inas Al Rahili and Arij Al‑Hadhli trained an AraBERT model to triage about 170,000 Arabic medical queries into three severity levels. The model achieved an F1 score of 98.29 %, outperforming a CNN baseline at 95.59 %.
Researchers also examined automated text recognition for Arabic manuscripts. Testing 500 lines of historical documents, one system had a 27.92 % error rate, while a specialized restoration model dropped errors to 4.06 %.
Legal‑tech attendees tested three language models on 90 bilingual legal questions about Saudi and EU data‑protection laws. The models showed 87 % consistency across 254 answer pairs, emphasizing the need for clearer compliance guidelines.
The conference underscored close ties with Malaysia’s national university, aiming to share AI expertise and joint research projects for Arabic‑speaking regions.
With these breakthroughs, Taibah University signals that AI tools tailored for Arabic contexts can compete with global systems, potentially transforming healthcare triage, education, and cultural heritage preservation in the region.