Half of Arabic Resumes Never Reach a Human Recruiter — Here's Why

If you're hiring in the GCC right now, there's a good chance your ATS is quietly filtering out half your Arabic-speaking candidates before a single human ever sees them. That's not a glitch — it's a structural flaw baked into the parsing engines that major enterprise platforms were built on. In this episode, we dig into a new analysis of over 840,000 Arabic resumes across MENA. The numbers are striking: Arabic CVs have an average 48% ATS pass rate, compared to 61% for English submissions globally. That 13-point gap is down to four compounding technical failures — right-to-left text reversal, Arabic's root-based morphology defeating keyword matching, font encoding errors, and dialect variation that most ATS models simply can't handle. We also look at what the fix looks like — from Arabic-native LLMs like Jais, ALLaM, and Fanar, to practical steps GCC employers can take right now to stop the talent bleed. And for candidates submitting into these systems, there's a counterintuitive workaround that might surprise you. This is AI HR Daily, brought to you by OVI — the AI-native ATS built for GCC hiring, with agents trained on Arabic CV parsing and local compliance frameworks.