Women Leading AI Development That Serves Arab Communities Daily

Building AI That Understands Arabic, Dialects and Identity

Fifty-four percent of Emirati students in MBZUAI’s 2026-27 cohort are women. That single figure captures something larger than enrollment statistics. It reflects a shift in who gets to shape the artificial intelligence systems that millions of people across the Arab world rely on every day to search, translate, communicate and learn.

The question at the center of that shift is deceptively simple: do these technologies actually understand the communities they serve? For the UAE, the answer has become urgent.

At Mohamed bin Zayed University of Artificial Intelligence (MBZUAI), researchers are confronting this directly. The work focuses on a technical problem with profound cultural stakes, teaching AI systems to recognize and process Arabic and its many regional dialects. The implications reach far beyond computation. Much of the UAE’s intangible heritage has traditionally been passed down orally. When AI systems fail to capture the nuance of spoken Emirati Arabic, they risk erasing cultural knowledge that has survived generations.

Dr. Hanan Aldarmaki leads this effort. Director of the Center for Teaching and Learning and Assistant Professor of Natural Language Processing at MBZUAI, she joined the university in 2022 as its first woman and first Emirati faculty member. Her research focuses on natural language and speech processing, with particular attention to Arabic and low-resource languages and dialects. That work extends through Ramsa Lab, a national initiative dedicated to documenting spoken Emirati dialects. The project generates datasets and resources that enable AI systems to better represent Emirati language and culture while preserving oral traditions that might otherwise be lost as technology reshapes how knowledge is stored and transmitted.

The technical challenge is real. Arabic presents a complex landscape for AI because of its many regional dialects. Spoken Emirati Arabic adds another layer of nuance that standard AI training data often misses. When systems cannot accurately process the way people actually speak, they fail users in real time. Translation becomes inaccurate. Digital assistants misunderstand requests. Search results miss relevant content. For speakers of minority dialects and low-resource languages, the gap between what AI can do and what it should do only widens.

Meanwhile, Aldarmaki’s own path reflects a broader shift taking place across the UAE. She completed postgraduate studies in the UK and the US before returning home to apply her expertise, holding an MPhil from the University of Cambridge and a PhD in Computer Science from George Washington University. Her return signals something significant: Emirati women are increasingly moving into scientific research and advanced technology roles, not simply participating in these sectors as users or professionals.

That transition is visible in the student body. The 54% figure in MBZUAI’s Emirati cohort reflects sustained investment in education, research and women’s participation in the workforce. For Aldarmaki, that ecosystem matters because it shapes who gets to decide what AI systems do next.

Her message to girls considering careers in AI is direct: build expertise. Understanding the language, concepts and technical foundations of a field gives women more than an entry point into technology. It gives them a voice in deciding where that technology goes. As the UAE positions AI as a pillar of its future economy, that distinction becomes consequential. Women who understand the field can shape whether AI systems serve all communities equitably or reinforce existing gaps.

Aldarmaki also resists a technology-first narrative. AI, she argues, is ultimately a tool, not the destination. The challenge for the next generation of researchers is not simply to build more capable systems but to ensure those systems contribute to human prosperity and a sustainable future. For Emirati women entering AI, that means having a role not just in adopting the technology but in deciding what its future looks like.

Whether the datasets being built today through projects like Ramsa Lab will be enough to close the gap for Arabic speakers remains an open question, and the answer will depend on who is sitting at the table when those decisions are made.