GCC AI News — 14 Oct 2026

The UAE ranks first worldwide for AI adoption, with 70.1% of its working-age population using AI, according to Microsoft's Global AI Diffusion Report highlighted by state news agency WAM around 11 October. The report names the UAE as the first economy anywhere to pass the 70% threshold, while much of the world remains below 10%. WAM noted AI has moved from an advanced technical tool to a core part of government services, education, healthcare and transport across the country, supported by national foundations built for resilience, continuity and responsible growth.

Adoption, not just investment, is what turns an AI strategy into everyday value. Leading the world on how many people actually use AI shows the UAE's long bet on skills, infrastructure and government services is paying off in real behaviour, not slideware.

A resident uses an AI assistant to renew a licence, a teacher plans lessons with AI, and a clinic triages patients faster — the 70% figure is these everyday uses added up across the population.

The Dubai Electronic Security Center built Saraab, an Emirati AI model that detects deepfake videos with up to 91% accuracy and can analyse clips as short as seven seconds — where comparable open-source models need at least 15. DESC says it is the first open-source deepfake-detection model developed by a government entity in the Arab region, built entirely by an Emirati team, and it scored 89% accuracy on data it had never seen in training. The model is being released on Hugging Face and GitHub so engineers, cybersecurity experts and researchers worldwide can test it, build on it and improve it.

Deepfakes threaten trust in everything from news to banking, and detection tools are scarce. A Gulf government not only building its own but open-sourcing it is a confident, generous move that helps defenders everywhere, not just at home.

A bank or newsroom runs a suspect video through Saraab to flag likely manipulation in seconds before acting on it, catching fakes that would slip past slower tools.

Qatar's Fanar 2.0, the second generation of its Arabic-centric generative AI platform from the Qatar Computing Research Institute at Hamad Bin Khalifa University, was built entirely in-country with no dependency on external AI providers. At its core is a 27-billion-parameter model with a 32K context window and native Arabic reasoning. Despite running on just 256 Nvidia H100 GPUs and using roughly eight times fewer training tokens than Fanar 1.0, it posted big benchmark gains — Arabic world knowledge up 9.1 points — plus new speech, vision, poetry, translation and safety components, all released openly on Hugging Face under an Apache 2.0 licence.

Arabic is only about 0.5% of web data despite 400 million speakers, so most models treat it as an afterthought. A sovereign, openly released Arabic stack from Qatar proves the Gulf can build culturally grounded AI for the region rather than importing it.

A Qatari ministry or startup builds an Arabic chatbot that understands Gulf dialect and cultural context on top of Fanar, instead of bolting Arabic onto an English-first model that mangles the nuance.

M42's National Reference Laboratory, with digital-pathology firm Qritive, has introduced AI-powered cancer diagnostics into its workflow at Cleveland Clinic Abu Dhabi. The system uses Qritive's AI Prostate module to analyse digitised biopsy slides, highlight suspicious regions, support detection of malignant glands and assist grading aligned with internationally recognised ISUP and Gleason criteria. Specialists stressed the AI is a support tool: it acts as a consistent 'second reader', but final diagnosis stays with clinicians. Doctors described it as a co-pilot that helps them focus expertise where it matters and review difficult cases faster.

Prostate-cancer treatment hinges entirely on how the disease is graded at diagnosis, a skilled task that can vary between reviewers. AI applying the same criteria every time makes diagnosis more consistent and timely, which directly shapes whether a patient gets surveillance, surgery or radiotherapy.

A pathologist at Cleveland Clinic Abu Dhabi reviews a prostate biopsy with the AI flagging areas of concern first, speeding up the read and reducing the chance a subtle pattern is missed.

HUMAIN, Saudi Arabia's PIF-backed national AI company, has brought its Horizon Ultra AI PC to market for enterprise buyers. Designed by its Edge Devices team in Riyadh and developed with Qualcomm over more than a year, the notebook runs on Snapdragon X2 Elite — an 18-core CPU with GPU and NPU — so AI workloads can run locally on the device and reach the cloud only when more power is needed. HUMAIN says on-device AI keeps sensitive data on the machine, supporting privacy and data sovereignty, and it pairs the hardware with its Arabic ALLaM models. The first generation targets one million AI PCs across the Middle East and Africa by 2030.

Most AI runs in distant data centres owned by a few global firms. A Gulf-designed AI PC that keeps data on the device gives regional governments and businesses a privacy-preserving, sovereign option — and signals Saudi Arabia moving up the stack from buying chips to designing devices.

A Saudi ministry handling sensitive citizen data runs AI document analysis directly on Horizon Ultra laptops, keeping records on-device instead of sending them to an overseas cloud.

HUMAIN has partnered with AI-agent company Napster to deploy Arabic-native AI agents across public and private organisations in Saudi Arabia and the wider Middle East. The partnership combines HUMAIN's ALLaM large language model, speech recognition and text-to-speech with Napster's video, voice and multimodal agent platform, with the aim of fielding Arabic-fluent 'digital employees'. It builds on HUMAIN's earlier moves to bring ALLaM to Microsoft Foundry and M365 Copilot, pushing Arabic-first agents into everyday business and government workflows across the region.

Most AI agents treat Arabic as an add-on to an English-first product, which breaks on dialect and cultural context. Pairing a sovereign Arabic model with a capable agent platform means Gulf organisations can deploy assistants that genuinely understand their customers and staff.

A Saudi utility fields an Arabic-speaking voice agent that handles billing questions in Gulf dialect end-to-end, instead of routing callers to a model that stumbles over local phrasing.

Abu Dhabi's M42 — a health group powered by AI, technology and genomics — is adopting Oracle Health Data Intelligence to unify clinical, genomic and real-world data across its network, advancing chronic-disease management, population health and precision care. Integrated with Oracle Health's EHR foundation, the platform aims to give clinicians and researchers faster insights and better care coordination. Built-in AI will help surface disease patterns, correlations and risk factors across large datasets, supporting earlier detection of emerging health trends that traditional analysis might miss. It builds on M42's existing work on unified longitudinal patient records across the UAE.

Healthcare value increasingly comes from connecting fragmented data, not just adding tools. A system that links clinical, genomic and real-world records across a national network lets AI find risks earlier — the difference between reacting to illness and preventing it.

A UAE care team spots a population-level rise in a chronic condition early through the unified platform, then targets screening and prevention before cases escalate into hospital admissions.

The Artificial Intelligence Office in the UAE government has launched a strategic collaboration with Dell Technologies to build national capabilities in agentic AI and move more applications from experiments into live government operations. The work covers building prototypes and implementation-ready solutions, specialised agentic-AI workshops for government entities, and access to the Dell AI Factory — an integrated stack of compute, software, data and services for training and deploying AI models securely at scale. It supports the UAE's target of shifting half of government operations to autonomous, agentic models within two years, and was formalised during a visit by over 50 UAE chief AI officers to the US.

The hard part of government AI is not demos but getting reliable, secure systems into daily operation. A partnership squarely aimed at crossing that pilot-to-production gap is how the UAE's agentic-AI ambitions become services citizens actually use.

A federal entity turns an agentic-AI prototype — say, one that processes permit applications end-to-end — into a secure production service, cutting waiting times for residents.

Abu Dhabi's Department of Government Enablement plans to launch Tamm 5.0, the next version of its government-services super-app, at GITEX Global in December, DGE director general Ruba Al Hassan told The National at the AI Everything summit. The update will make services 'increasingly intelligent' and more proactive and personalised, building on the Tamm AI Assistant that already completes many services on a user's behalf, plus AI Vision, a Smart Guide for document scanning, Arabic-English voice and translation, and personalised dashboards that flag deadlines. It is part of Abu Dhabi's goal of becoming an AI-native government, redesigning public services around AI from the ground up.

A government super-app that anticipates needs rather than waiting for forms is a concrete example of AI improving daily life. Abu Dhabi redesigning services around AI from the ground up sets a template other governments can study.

A resident opens Tamm and is proactively reminded of an upcoming renewal, with the AI assistant completing most of the paperwork on their behalf instead of them navigating multiple departments.

Saudi Arabia is consolidating its position as a global data and AI hub under Vision 2030, having designated 2026 the Year of Artificial Intelligence, according to the Saudi Press Agency. Led by the Saudi Data and AI Authority, the national strategy spans ambition, skills, policy, investment, innovation and ecosystem. Government spending on emerging technologies rose more than 56% in 2024, AI companies secured $9.1 billion in funding, over 11,000 specialists have been trained, and the SAMAI programme has reached more than one million participants. The Kingdom ranked 14th in the 2025 Global AI Index and leads the Arab world on AI model development.

Turning a national vision into trained people and funded companies is what separates AI rhetoric from capability. Saudi Arabia pairing a 'Year of AI' with real spending, training at scale and sovereign infrastructure shows a long-term build, not a one-off announcement.

A young Saudi graduate trained through SAMAI joins a local AI company funded under the strategy, helping build Arabic-first products instead of leaving the region for opportunities abroad.