GCC AI News — 5 Oct 2026

HUMAIN, the Public Investment Fund-owned full-stack AI company, announced on October 4 that it has engaged professional-services firm EY to develop Saudi Arabia's first AI-native business process outsourcing service, underpinned by HUMAIN's ecosystem. The service aims to create a new category of AI-enabled business services that combine intelligent agents, automation, advanced analytics and human expertise into integrated operating models. CEO Tareq Amin said that with HUMAIN ONE, 'we are turning AI from a tool into the operating layer of the enterprise, built and deployed at scale from Saudi Arabia to the world.' EY will handle commercialization and industry expertise while HUMAIN supplies the underlying AI technology and agentic infrastructure.

This positions Saudi Arabia not just as a buyer of AI but as an exporter of AI-run services. Rebuilding outsourcing around intelligent agents could leapfrog the Kingdom past the traditional labor-heavy BPO model and create a template other Gulf economies can follow.

A regional enterprise can engage HUMAIN ONE to run finance, procurement or customer operations as AI-native workflows — getting agent-plus-human service delivery without building its own automation stack in-house.

Ai Everything Abu Dhabi 2026 opens October 6-7 at ADNEC, hosted by Abu Dhabi's Department of Government Enablement with over 400 AI-first companies, 180-plus speakers and 200 investors representing $91bn. The summit showcases Abu Dhabi's $3.54bn programme to become the world's first AI-native government by 2027, linking more than 40 government entities into a unified digital ecosystem with over 100 AI use cases already live in public services. Separately, UAE AI Minister Omar Sultan Al Olama said in late September the country aims to move 50% of government services and operations to agentic AI models within two years. A Microsoft Copilot rollout is reaching 35,000 Abu Dhabi government employees.

The UAE is attempting something few governments have tried: making AI the default operating model for the state, not a pilot. If it works, citizens get faster services and the UAE becomes a live reference other governments study and copy.

A GovTech vendor or systems integrator can use Ai Everything as the entry point to Abu Dhabi's 100-plus live AI use cases, finding concrete public-sector deployments to partner on rather than pitching abstract capability.

HUMAIN has unveiled humain-m3, a frontier Arabic-language model now available as a research and evaluation preview on its HUMAIN Node developer platform. The 428-billion-parameter mixture-of-experts system was built in partnership with AI firm MiniMax, with HUMAIN adding pre-training on more than a trillion tokens of Arabic-native text for regional specialization. The release is part of Saudi Arabia's push to strengthen its Arabic AI ecosystem, giving developers and enterprises access to advanced inference tuned for the language and cultural context, with a planned open-weight release to broaden access for the regional developer community.

Most leading models treat Arabic as an afterthought. A high-performance, Arabic-first model trained on native text can serve 400 million Arabic speakers far better — unlocking government, education and customer-service applications that generic models handle poorly.

A regional developer building an Arabic chatbot or document tool can prototype on humain-m3 via HUMAIN Node, getting stronger Arabic comprehension and dialect handling than retrofitted English-first models deliver.

The UAE Cabinet, chaired by Sheikh Mohammed bin Rashid Al Maktoum at Qasr Al Watan, approved the launch of the Cabinet AI Advisor: a system of 32 specialised AI advisors that analyse policies and legislation, assess financial, economic, social and environmental impact, compare them with global best practice and present recommendations to ministers around the clock. At the same September meeting, the Cabinet approved rolling an AI curriculum out to all public and private schools after a public-school pilot, and reviewed a programme to train 22,000 teachers to use AI in teaching, assessment and lesson planning. Officials stressed the advisors support ministers' decisions rather than replace them.

This pairs top-down adoption with bottom-up capability: AI helps run the government while the next generation learns to build and question it. Training 22,000 teachers and reaching every school is how a country turns AI from an elite tool into a national skill.

An edtech company can align its product with the UAE's seven-area AI curriculum framework and teacher-training needs, serving a nationwide rollout rather than selling school by school.

Qatar's Hamad Medical Corporation, the country's main public healthcare provider, has gone live with an Arabic clinical AI platform across its nine Doha hospitals, built on Google Cloud's Vertex AI and a Qatar Computing Research Institute fine-tuned foundation model. The system handles radiology report drafting, Arabic clinical note generation and patient-communication drafting, with consultants reviewing and signing off every output. Internal pilot data showed roughly a 45% reduction in radiology first-draft time and a 30% cut in note-typing load. The deployment makes Qatar the first Gulf state with a production Arabic clinical large language model running at national hospital scale, integrated with the national electronic medical record.

This is one of the region's first proofs that Arabic medical AI works in live care, not demos — freeing clinicians to spend more time with patients and making records clearer. Mandatory human sign-off keeps doctors in control, a model other Gulf health systems can safely copy.

A radiology or clinical department can adopt a review-and-sign-off AI drafting workflow like HMC's to cut first-draft time dramatically while keeping a qualified clinician accountable for every report.

The UAE Ministry of Justice showcased a suite of five specialised AI agents that cut the time to review case files from several days to a few hours, demonstrated on the sidelines of the 15th UN Congress on Crime Prevention and Criminal Justice at ADNEC, Abu Dhabi in late September. The agents integrate with the Ministry's electronic court systems: one classifies documents and extracts information, another links facts and identifies legal issues, a third compiles relevant legislation and precedents, a fourth summarises options for the judge, and a fifth prepares a non-binding preliminary draft judgment. Of seven stages in the review process, five are AI-supported while two remain exclusively with the judge, who reviews and approves all output.

This shows AI accelerating public services without ceding judgment: the technology does the heavy lifting of research and drafting while final decisions stay human. Faster case handling means citizens wait less for justice — a tangible, everyday benefit of responsible AI adoption.

A legal team or court can adopt an AI-assisted pipeline that extracts facts and drafts preliminary analysis, compressing routine review work while keeping a qualified human responsible for every binding decision.

Bahrain's Labour Fund (Tamkeen) launched an enhanced Digital Enablement and AI Programme in early October to accelerate digital transformation among Bahraini enterprises, especially micro, small and medium-sized businesses. Under the scheme, Tamkeen covers up to 50% of eligible costs for software solutions and related services, giving firms flexibility to select and implement AI and digital tools matched to their operations. The programme builds on Tamkeen's earlier Digital Enablement initiative and reflects its 2026 strategic priorities of high-impact economic growth, private-sector development and creating career opportunities for Bahrainis by aligning skills with market demand.

Small businesses are usually the last to afford AI. By subsidising half the cost, Bahrain lowers the barrier for the firms that employ most of its workforce — spreading AI's productivity gains across the real economy rather than concentrating them in large corporations.

A Bahraini SME can apply for Tamkeen support to adopt an AI tool — customer service, bookkeeping or marketing automation — recovering up to half the cost and testing real productivity gains at reduced risk.

Oman is expanding digital-skills training for young Omanis through Ericsson's Digital Lab programme, giving students hands-on experience in AI, robotics, coding, electronics and other emerging technologies, as reported October 2. The initiative is part of Oman's broader push under Vision 2040 to develop national talent in advanced technology, complementing the Sultanate's moves to build a semiconductor ecosystem and an AI Special Zone in Muscat established by royal decree. The programme focuses on practical, lab-based learning designed to prepare Omani youth for careers in the technologies reshaping the regional economy.

Oman is investing in people before pouring concrete, recognising that AI infrastructure is only as useful as the engineers who run it. Hands-on training for young nationals builds a durable domestic talent base — the foundation of any sustainable tech economy.

A university or employer in Oman can partner with a Digital Lab-style programme to give students and staff practical AI and robotics experience, closing the gap between classroom theory and the skills regional employers actually need.

Kuwait's Central Agency for Information Technology (CAIT) developed 'Wasel,' a unified AI-powered platform built to manage the citizen and resident experience across government services. CAIT describes Wasel as national digital infrastructure rather than a single project, relying on AI-driven smart automation, real-time data analysis and automated classification and routing to handle requests. The effort is part of Kuwait's drive under Vision 2035 to digitalise more than 90% of government services, and builds on CAIT's partnership with Microsoft to establish an AI-powered Azure region and advance local AI capabilities across the public sector.

A single AI-driven gateway can replace the maze of separate portals citizens usually navigate, cutting waiting times and errors. For Kuwait, it is a concrete step from strategy documents to services people actually use — the hardest part of digital government.

A government service team can model a unified intake platform on Wasel — using AI to classify and route citizen requests automatically — reducing manual triage and getting residents to the right service faster.

MBZUAI's Institute of Foundation Models released K2 Horizon, a family of six models ranging from 0.9 billion to 375 billion parameters, publishing the weights, code, training data and methodology. The fully open release, part of a wave of UAE AI activity in early autumn, lets researchers and developers inspect, reproduce and build on the models freely rather than treating them as closed black boxes. It reinforces the UAE's positioning as a producer of open, sovereign AI and a contributor to the global research community, alongside Abu Dhabi's broader push into government AI and advanced-model research.

Fully open models — not just open weights — are rare and valuable: they let anyone audit how a model was built and adapt it for their own language, domain or region. The UAE backing genuine openness strengthens trust and gives the Global South capable models it can actually control.

A researcher or startup can download a K2 Horizon model and fine-tune it on local data with full visibility into its training, building a domain-specific system without paying for or depending on a closed commercial API.