Morocco Hosts AbjadNLP 2026: Rabat Workshop Advances Arabic Script NLP Amid MENA Localisation Surge
Rabat, Morocco, will host the AbjadNLP 2026 workshop on Saturday, 28 March 2026. This full-day event focuses on natural language processing for languages using Arabic script. It builds on growing MENA efforts to localise AI for Arabic dialects and heritage texts. Organisers expect it to draw researchers tackling core challenges in Arabic NLP.ArabicNLP 2026
The workshop comes as Arabic-specific models like Jais and AceGPT gain traction across the Gulf and North Africa. Enterprise tools for Arabic still lag English by 20-30% in performance. Events like this aim to close that gap through shared benchmarks and dialect support.
Rabat Emerges as Arabic AI Hub
Morocco positions itself as a key player in Arabic NLP research. The AbjadNLP 2026 event, set for 9am to 5:45pm in Rabat, targets languages with Arabic script, including Arabic, Persian, Urdu, and Pashto. It follows the first edition's success and coincides with broader conferences like EACL 2026.AbjadNLP 2026
Rabat's choice underscores North Africa's rising role. The city hosted planning for the Arabic NLP School 2026 on 24 March, co-located with EACL. Over 100 attended the prior school in Abu Dhabi. This momentum supports MENA-wide localisation, from UAE voice agents to Saudi LLMs.Arabic NLP School 2026
Arabic is genuinely hard for computers. Large language models have gotten significantly better at Arabic, but dialect support is still weak.
Local efforts align with models like Jais from UAE's Inception and ALLaM from Saudi's SDAIA. These tools handle Modern Standard Arabic well but falter on dialects.
Tackling Arabic NLP Challenges
Arabic NLP faces unique hurdles: complex morphology, right-to-left script, and dialect variation. Models like Claude and GPT-4 now manage nuanced conversations, yet enterprise applications underperform. Segmentation, POS tagging, and NER lag due to limited benchmarks.Arabic NLP 2026 Assessment
The Fourth ArabicNLP Conference calls for papers on core tasks: tokenisation, parsing, and SRL. Deadlines approach for submissions advancing these areas.ArabicNLP 2026 CFP
In MENA, localisation drives progress. UAE's Tarjama unveiled next-gen LLMs at MenaML26, as covered in Tarjama CEO Nour Al Hassan Unveils Next-Gen Arabic LLMs at MenaML26 in UAE and Jordan. Gulf voice agents from Arabic.AI and HeyBreez Have Gone Live Inside Etisalat show commercial viability.
Workshops like AbjadNLP foster open-source tools. Libraries such as CAMeL Tools and AraBERT provide foundations. Dialect benchmarks remain scarce, pushing events to create Arabic MMLU equivalents.
Manuscript AI and Shared Tasks
A standout initiative is NAKBA NLP 2026, a shared task on Arabic manuscript understanding. It uses the Omar Al-Saleh Memoir Collection for OCR and transcription. Two tracks compete: manual transcription and automatic models.NAKBA NLP 2026
Teams develop systems on CodaBench. High-resolution images test OCR accuracy for heritage texts. This aids digitisation efforts in MENA archives, from Moroccan libraries to Saudi manuscripts.
Such tasks mirror MENA trends. Egypt's KARNAK LLM (recently launched) boosts Cairo's AI status. Qatar's Fanar excels in clinical models. Morocco's focus complements these, emphasising script-based NLP.
- Transcription Track: Enrich corpus with manual transcriptions for unseen pages.
- Systems Track: Build and evaluate automatic models for Arabic manuscripts.
- Benchmark: Open dataset of manuscript images for OCR research.
- Goal: Advance heritage preservation via AI.
By The Numbers
- 2026 AbjadNLP 2026 runs a full day from 9am to 5:45pm in Rabat, expecting 100+ attendees based on prior Arabic NLP School turnout in Abu Dhabi.Arabic NLP School 2026
- 20 Enterprise Arabic NLP tools perform 20-30% worse than English counterparts in sentiment analysis and document processing.Arabic NLP 2026
- 4 Jais and AceGPT models show dramatic gains, with GPT-4 handling 85% accurate Arabic translations per internal benchmarks.alsahli.dev
- 500 m NAKBA NLP provides datasets from over 500 manuscript pages, targeting 90% OCR accuracy in shared tasks.NAKBA NLP
- $1.2 billion MENA Arabic LLM investments hit $1.2 billion in 2025, per SDAIA reports, fueling localisation.SDAIA
| Model | Developer | Strength | MENA Impact |
|---|
| Jais | Inception | MSA + Dialects | UAE Enterprise Chat |
| ALLaM | SDAIA | Long-Context Arabic | Saudi Gov Apps |
| AceGPT | Open-Source | Fine-Tuning | Regional Benchmarks |
| Fanar | Qatar | Clinical NLP | Healthcare Localisation |
Evaluation benchmarks for Arabic are limited. We need Arabic equivalents of MMLU, HellaSwag, and other standard benchmarks.
MENA Localisation Momentum
MENA nations prioritise Arabic AI. UAE leads with MBZUAI training Falcon models. Saudi's PIF funds infrastructure, linking to Saudi MIS Signs $501 Million AI Data Centre Contract.
Morocco's Rabat events signal North African entry. Technology Innovation Institute in Abu Dhabi shares tools via Falcon. Partnerships with Microsoft and Nvidia accelerate progress.
Challenges persist: dialects vary across 22 Arab countries. Workshops address this via collaborative benchmarks. Outcomes will feed into production LLMs for customer service and education.
Regional context
The MENA AI market is compounding at a pace that few advanced-economy benchmarks now match. IDC forecasts the region will cross $15 billion in AI spend in 2026, up from $8.7 billion in 2024, and PwC projects a $320 billion contribution to Gulf GDP from AI by 2030. Sovereign capital is the dominant funder, with PIF, Mubadala, and the Qatar Investment Authority all disclosing increased AI allocations over the last six months. Against that backdrop, the Morocco story we cover here sits inside a broader pattern: Gulf governments and their partners in Cairo, Rabat, and Amman are translating national strategies into procurement, and the procurement decisions are starting to favour Arabic-first vendors who can show production deployments rather than demos.
Regional context
The MENA AI market is compounding at a pace that few advanced-economy benchmarks now match. IDC forecasts the region will cross $15 billion in AI spend in 2026, up from $8.7 billion in 2024, and PwC projects a $320 billion contribution to Gulf GDP from AI by 2030. Sovereign capital is the dominant funder, with PIF, Mubadala, and the Qatar Investment Authority all disclosing increased AI allocations over the last six months.
Against that backdrop, the Morocco story we cover here sits inside a broader pattern. Gulf governments and their partners in Cairo, Rabat, and Amman are translating national strategies into procurement, and those procurement decisions are starting to favour Arabic-first vendors who can show production deployments rather than demos. Gartner notes that MENA public-sector AI contracts crossed 400 distinct awards in the 12 months to March 2026, roughly double the figure for the same period in 2024. Private demand is tracking that growth across financial services, healthcare, and logistics, with Deloitte reporting that 63% of large GCC employers now treat AI capability as a board-level performance indicator, up from 34% in 2023.
The AI in Arabia View: Morocco's AbjadNLP 2026 cements Rabat as a North African AI node, bridging Gulf LLM giants like Jais with heritage-focused tasks like NAKBA. This script-centric approach tackles Arabic's morphology head-on, where English models falter by 25%. Expect dialect datasets to emerge, powering localised apps from Casablanca call centres to Riyadh archives. MENA's $2 billion AI spend in 2026 demands such events; without them, localisation stalls at MSA. Rabat's play elevates Maghreb voices in a UAE-Saudi dominated field, fostering pan-Arab benchmarks that could standardise evaluation by 2027.
