G42 — UAE AI and Cloud Computing Powerhouse

G42 stands as the UAE’s most significant contribution to the global AI landscape. Backed by a 2.3 billion dollar investment from Microsoft in 2024 and operating the Condor Galaxy supercomputer through its partnership with Cerebras Systems, G42 has positioned itself as the infrastructure and model development hub for Arabic artificial intelligence.

The company’s AI strategy operates across three pillars. First, foundation model development through the Jais LLM family, produced by G42’s Inception unit in partnership with MBZUAI and Cerebras. Second, cloud infrastructure providing the computing fabric that Arabic AI applications require for training and inference. Third, enterprise AI solutions that package G42’s capabilities into products for government, healthcare, energy, and financial services customers.

Microsoft’s investment was both financial and strategic. The partnership provides G42 with access to Azure cloud infrastructure, enterprise distribution channels, and Microsoft’s global customer base. In return, Microsoft gains a foothold in the Gulf AI market and a partner capable of delivering Arabic AI capabilities that Microsoft’s own models cannot match. The Jais models are deployed on Azure alongside Microsoft’s own offerings, creating a unique catalog that serves both English and Arabic enterprise customers.

G42’s Inception unit functions as the company’s AI research and model development arm. Beyond Jais, Inception conducts research in computer vision, cybersecurity, and domain-specific AI applications. The unit’s access to the Condor Galaxy supercomputer — one of the largest non-hyperscaler AI training systems globally — provides computational resources that enable frontier model development independent of US or Chinese cloud providers.

The company’s revenue model combines B2B cloud services, enterprise AI solutions, and government contracts. G42’s relationship with the UAE government provides both a large customer base and strategic direction, with the company executing on national AI priorities while maintaining commercial operations that generate independent revenue.

Jais LLM Development Program

G42’s most consequential technical contribution is the Jais large language model family, developed through the trilateral partnership between G42’s Inception unit, MBZUAI, and Cerebras Systems. This collaboration unites Emirati sovereign ambition, academic research excellence, and American supercomputing hardware into a program that has fundamentally altered the competitive landscape of Arabic AI.

The development timeline spans four generations. Jais-13B launched in August 2023 with 13 billion parameters trained on 116 billion Arabic tokens and 279 billion English tokens. Jais-30B followed three months later with doubled capacity. The Jais Family Release in H2 2024 published 20 open-source Arabic-centric models spanning 590 million to 70 billion parameters, trained on up to 1.6 trillion tokens — the largest single model release in MENA history. Jais 2, arriving in December 2025 with 70 billion parameters trained on 600+ billion Arabic tokens, established the model as the world’s most advanced Arabic open-weight LLM.

Jais 2 covers Modern Standard Arabic and 17 regional dialects including Gulf varieties (UAE, Saudi, Kuwaiti, Bahraini, Qatari, Omani), Egyptian Arabic, Levantine varieties (Palestinian, Jordanian, Lebanese, Syrian), Iraqi Arabic, Maghrebi varieties (Moroccan, Algerian, Tunisian, Libyan), and Sudanese Arabic. The model also processes Arabizi, the romanized Arabic writing system dominant in informal digital communication among younger Arabic speakers.

Condor Galaxy Computing Infrastructure

The Condor Galaxy 1 supercomputer, built jointly with Cerebras Systems, provides G42 with sovereign AI training capability. Based on Cerebras CS-2 wafer-scale engines, each integrating 850,000 AI-optimized compute cores onto a single chip, the system eliminates the communication bottlenecks that limit conventional GPU cluster efficiency. The multi-exaFLOP performance enables training campaigns — like Jais 2’s 70B parameters on 600+ billion Arabic tokens — within commercially viable timeframes.

The wafer-scale architecture provides specific advantages for Arabic LLM training. Arabic’s morphological complexity means equivalent semantic content requires more tokens than English, increasing training sequence lengths. The CS-2’s on-chip memory bandwidth maintains high utilization at these longer sequences, whereas GPU clusters experience degraded efficiency. The training cost per token is lower than conventional GPU infrastructure, making iterative Arabic model development economically feasible.

G42 has reported that ML techniques uniquely enabled by Cerebras hardware achieved state-of-the-art quality using only a fraction of the compute used to train similar-sized models in the past — an efficiency claim that, if validated at scale, positions Cerebras-based infrastructure as the cost-effective choice for Arabic foundation model development.

Competitive Positioning Against HUMAIN and TII

G42 competes in a three-way race with HUMAIN (Saudi Arabia) and TII (Abu Dhabi) for Arabic AI leadership. HUMAIN, launched in May 2025 with $77 billion in planned data center investment and deals exceeding $23 billion since launch, pursues scale through massive infrastructure. The xAI partnership (500 MW data center), Adobe tenancy, and partnerships with NVIDIA, AMD, and AWS create ecosystem breadth that G42’s more focused model-development approach does not match in infrastructure terms.

TII, through Falcon-H1 Arabic, leads the Open Arabic LLM Leaderboard at 75.36 percent with a hybrid Mamba-Transformer architecture. The architectural innovation — combining Mamba state-space models with transformer attention — delivers performance at parameter counts that pure transformer models (including Jais 2) cannot match at equivalent computational cost. Falcon-H1’s 256,000-token context window exceeds what Jais 2 can offer with pure transformer architecture.

G42’s competitive advantages lie in three areas. First, the Cerebras partnership provides unique computing infrastructure that neither HUMAIN’s GPU-based data centers nor TII’s research clusters can replicate for training efficiency. Second, the Microsoft partnership ($2.3 billion investment, Azure deployment) provides enterprise distribution channels and global market access. Third, the Jais model family’s four-generation development lineage provides accumulated training expertise that newer programs lack.

Market Context

G42 operates within a MENA AI ecosystem experiencing rapid growth. AI-focused venture capital reached $858 million in 2025, representing 22 percent of total VC funding. The UAE AI market is projected to grow from $578 million in 2024 to $4.25 billion by 2033 at a 22.07 percent CAGR. The UAE attracted $519 million in AI funding in 2025, confirming its position as the primary AI hub in the Gulf.

The OpenAI-G42 Stargate project — a planned 1 GW AI computing cluster in Abu Dhabi — would represent the largest single AI infrastructure deployment in the MENA region. Combined with HUMAIN’s Saudi data center buildout, TII’s research infrastructure, and the broader startup ecosystem, the Gulf is developing AI computing capacity that collectively represents the largest investment outside the United States and China.

G42’s position within the broader UAE AI strategy includes alignment with national priorities articulated in the UAE AI Strategy. The strategy targets AI contribution to 12 percent of GDP by 2031, with G42 serving as the primary vehicle for commercial AI capability development. The company’s dual role — executing sovereign AI priorities while generating commercial revenue — creates both advantages (government customer base, strategic direction) and constraints (alignment with national policy requirements, geopolitical considerations affecting international partnerships).

G42’s AI Product Portfolio Beyond Jais

G42’s AI portfolio extends beyond the Jais LLM to encompass cloud computing, computer vision, cybersecurity, and healthcare AI. The company’s AI42 cloud platform provides Arabic-first cloud services for enterprise customers across the MENA region. Computer vision applications serve UAE government clients in surveillance, traffic management, and urban planning. Healthcare AI deployments, developed through the Omais subsidiary, provide diagnostic support and clinical decision assistance in Arabic-language healthcare settings.

The $2.3 billion Microsoft investment in G42 in 2024 reshaped the company’s commercial trajectory. The partnership integrates G42’s sovereign AI capabilities — Jais models, Condor Galaxy computing, Arabic-specific AI applications — with Microsoft’s Azure cloud ecosystem. Jais deployment on Azure provides global distribution while maintaining training sovereignty on Condor Galaxy. The partnership creates a sovereign-train, global-deploy model that other nations studying AI sovereignty strategies are evaluating.

G42’s venture investment activity complements its internal AI development. Investments across the MENA AI startup ecosystem position G42 as both a platform provider (Jais LLM, AI42 cloud) and an ecosystem investor, creating multiple channels through which G42 captures value from Arabic AI growth. The Stargate UAE project — a planned 1 GW AI computing cluster in Abu Dhabi in partnership with OpenAI — extends G42’s infrastructure position beyond the Condor Galaxy system, ensuring computing capacity for the next generation of Arabic AI development.

The broader competitive dynamics between G42 (UAE) and HUMAIN (Saudi Arabia) reflect the Gulf states’ parallel investments in AI sovereignty. G42’s advantages — the Cerebras-based Condor Galaxy for training efficiency, the Microsoft partnership for global distribution, and four generations of Jais development experience — complement its position as the UAE’s primary AI champion. The $858 million in MENA AI VC during 2025 flows through an ecosystem where G42 serves as both a corporate investor and a platform provider for Arabic AI startups building on Jais and G42 cloud infrastructure.

G42’s Strategic Positioning and Future Trajectory

G42’s trajectory from a UAE-focused AI company to a globally integrated AI platform reflects the maturation of Gulf AI ambitions. The Microsoft partnership provides global distribution infrastructure, Azure integration creates enterprise deployment pathways, and the Condor Galaxy supercomputer provides sovereign training capability. This combination — sovereign training with global distribution — establishes a model that other nations studying AI sovereignty strategies are evaluating for adaptation.

The competitive dynamics between G42 and HUMAIN reflect broader UAE-Saudi technology rivalry that benefits the Arabic AI ecosystem through accelerated development. G42’s Jais 2 and HUMAIN’s ALLaM 34B provide Arabic speakers with two competitive sovereign LLM options, each with distinct strengths (Jais for dialect breadth, ALLaM for institutional knowledge). The competition ensures continued investment and innovation in Arabic AI, preventing the complacency that monopoly positions can create.

G42’s participation in the Stargate UAE project alongside OpenAI signals expanding partnerships with leading US AI companies. This partnership positions the UAE as a bridge between American AI innovation and MENA market deployment — a strategic role that G42 executes through its dual identity as both a sovereign AI champion and a Microsoft partner with global ambitions. The $858 million in MENA AI VC during 2025 flows through an ecosystem where G42 serves as anchor investor, platform provider, and model developer — creating multilateral network effects that strengthen the UAE’s position as the commercial hub of Arabic AI development.

G42’s evolution from a UAE-focused AI company to a globally integrated platform reflects the maturation of Gulf AI ambitions into commercially viable international operations. The combination of sovereign computing (Condor Galaxy), partnership distribution (Microsoft Azure), open-weight models (Jais), and ecosystem investment creates a multi-layered platform that generates value across the Arabic AI ecosystem while maintaining the sovereign capabilities that the UAE’s national AI strategy requires. The Stargate UAE project’s 1 GW computing cluster ensures that G42’s infrastructure advantage will scale with the next generation of Arabic AI development.

As the MENA AI market grows — $858 million in AI VC during 2025, UAE market projected to $4.25 billion by 2033 — G42’s platform position provides both direct revenue from AI services and indirect returns from ecosystem investments in startups building on Jais and G42 cloud infrastructure. This dual-revenue model sustains the investment intensity that Arabic AI development requires while generating commercial returns that validate the sovereign AI development thesis.

Jais Model Family as Commercial Platform

G42’s commercial strategy centers on the Jais model family as a platform product rather than a standalone model release. The 2024 release of 20 open-source models ranging from 590 million to 70 billion parameters — the largest single model release in MENA — provides customers with Arabic LLM options across the full compute spectrum. Jais 2 (70B, December 2025) with 600 billion Arabic training tokens and 17 dialect coverage represents the flagship capability. Available through Hugging Face and JaisChat.ai, the Jais family enables deployments from edge devices (sub-7B models) through enterprise GPU servers (70B model).

This platform approach generates value through cloud services, consulting, and ecosystem investment rather than model licensing fees. Organizations deploying Jais on G42 cloud infrastructure pay for compute and support rather than model access, creating a recurring revenue model tied to deployment scale. The open-weight availability ensures broad adoption and community contribution that accelerates model improvement through feedback, fine-tuning discoveries, and deployment best practices shared across the 180+ organizations that submit models to the OALL.

Related Coverage

• MENA AI Companies — Full company directory
• Arabic LLMs — Foundation model coverage
• Jais — Arabic LLM — Model family analysis
• Condor Galaxy — Supercomputer infrastructure
• MBZUAI Profile — Research partnership
• HUMAIN Profile — Saudi competitor
• TII Profile — Abu Dhabi competitor
• MENA Startup Ecosystem — Investment landscape