Amazon Web Services (AWS)
In the UAEThe largest cloud provider by market share, and the infrastructure most AI teams still default to first - GPU instances, managed model access through Bedrock, and the storage and networking underneath both.
What does AI actually run on?
A stack of 8 layers, from the person asking down to the power socket - pick one to see who runs it.
ComputeThe processing power
The raw computing capacity - rented by the hour, the month or the year - that trains and runs AI models. This is what "the cloud" mostly means in practice.
See the 10 providersUser
The person
Where everything starts: someone typing a question, using a search box, or opening an app that happens to have AI built into it. Nothing below this layer matters unless it eventually reaches a real person doing something with it.
The technical version
The human, or the downstream system, issuing the request. Not itself infrastructure - but the layer every request originates from and every response is optimised to return to, fast enough to feel instant.
This layer sits on top of the infrastructure rather than being part of it, so it is covered in another part of Nakoda.
See the AI products people actually useApplication
The product
The app, website or feature that wraps an AI model into something usable - a chatbot, a copilot inside other software, a search assistant. This is the layer most people actually mean when they say "AI."
The technical version
The product layer: orchestration, prompting, business logic, guardrails and interface sit here, calling one or more models beneath it rather than running any model itself.
This layer sits on top of the infrastructure rather than being part of it, so it is covered in another part of Nakoda.
Browse the companies building AI applicationsModel
5 providers at this layer
The trained system that actually produces the answer - GPT, Gemini, Claude, Llama and others like them - plus the platforms that make models like these available for developers to build on.
The technical version
Foundation models and the model-hosting layer beneath them: managed access to first- and third-party models (Bedrock, Azure OpenAI Service, Vertex AI), or open-weight model deployment (Hugging Face, Together AI).
The open-source hub most of the AI industry builds on: a repository of models and datasets, plus a managed way to deploy any of them - Hugging Face's own or the open-weight models it hosts - to production.
Microsoft Azure
In the UAEMicrosoft's cloud platform, and the first of the major clouds to open a dedicated Gulf region. Azure OpenAI Service is the most common route enterprises use to buy OpenAI's models under their own compliance terms.
Google Cloud
Not in the UAEGoogle's cloud platform and home of the Gemini model family through Vertex AI. Notably absent from the UAE specifically: its nearest Gulf regions are in Qatar and Saudi Arabia, not the Emirates.
An "AI-native cloud" built specifically for open-source models: serverless, pay-per-token inference on 200+ models, plus dedicated GPU clusters for teams that want to train or fine-tune their own.
Inference
7 providers at this layer
The moment a trained model is actually used to produce an answer, in about the time it takes to read this sentence. Training a model happens once; inference happens every single time someone uses it.
The technical version
Serving pre-trained model weights against live requests, optimised for latency and cost per token rather than for training throughput - the domain of purpose-built inference chips, edge inference and serverless model endpoints.
Amazon Web Services (AWS)
In the UAEThe largest cloud provider by market share, and the infrastructure most AI teams still default to first - GPU instances, managed model access through Bedrock, and the storage and networking underneath both.
The open-source hub most of the AI industry builds on: a repository of models and datasets, plus a managed way to deploy any of them - Hugging Face's own or the open-weight models it hosts - to production.
Microsoft Azure
In the UAEMicrosoft's cloud platform, and the first of the major clouds to open a dedicated Gulf region. Azure OpenAI Service is the most common route enterprises use to buy OpenAI's models under their own compliance terms.
Cloudflare
In the UAEBest known as a network and security company, Cloudflare also runs GPU inference at the edge through Workers AI - and has operated a Dubai point of presence since 2018, well before most AI-specific providers reached the region.
Google Cloud
Not in the UAEGoogle's cloud platform and home of the Gemini model family through Vertex AI. Notably absent from the UAE specifically: its nearest Gulf regions are in Qatar and Saudi Arabia, not the Emirates.
Groq
Not in the UAEGroq builds LPUs - chips designed specifically for inference rather than training - and sells access to them through GroqCloud. Its major Gulf build-out is in Saudi Arabia, not the UAE.
An "AI-native cloud" built specifically for open-source models: serverless, pay-per-token inference on 200+ models, plus dedicated GPU clusters for teams that want to train or fine-tune their own.
Compute
10 providers at this layer
The raw computing capacity - rented by the hour, the month or the year - that trains and runs AI models. This is what "the cloud" mostly means in practice.
The technical version
Virtualised and bare-metal compute: CPU and GPU instances, managed clusters, HPC-grade interconnect and reserved capacity, provisioned by hyperscale clouds and specialist GPU clouds alike.
Amazon Web Services (AWS)
In the UAEThe largest cloud provider by market share, and the infrastructure most AI teams still default to first - GPU instances, managed model access through Bedrock, and the storage and networking underneath both.
e& enterprise
UAE-basedThe enterprise technology arm of e& (formerly Etisalat), and the operator behind OneCloud - a sovereign hyperscale cloud built with Oracle, hosted and governed entirely inside the UAE.
G42 / Core42
UAE-basedAbu Dhabi's flagship AI technology group. Core42, its cloud and AI-infrastructure company, runs sovereign AI cloud services on NVIDIA GPUs; G42 co-developed the Condor Galaxy AI supercomputer network with Cerebras. The closest thing the UAE has to its own hyperscaler.
Microsoft Azure
In the UAEMicrosoft's cloud platform, and the first of the major clouds to open a dedicated Gulf region. Azure OpenAI Service is the most common route enterprises use to buy OpenAI's models under their own compliance terms.
NVIDIA
Not in the UAEThe company whose chips sit under almost all of this stack. NVIDIA does not operate UAE infrastructure itself, but its GPUs are what every cloud and sovereign-AI provider in this catalog is actually renting out.
Oracle Cloud Infrastructure (OCI)
In the UAEOracle's cloud platform, and the operator of two UAE regions - Dubai and Abu Dhabi - the latter now home to the Middle East's first OCI Supercluster built on NVIDIA Blackwell GPUs for sovereign AI workloads.
A GPU-specialist cloud that grew out of crypto mining infrastructure into one of the largest independent NVIDIA GPU providers, built specifically for AI training and inference rather than general-purpose computing.
Google Cloud
Not in the UAEGoogle's cloud platform and home of the Gemini model family through Vertex AI. Notably absent from the UAE specifically: its nearest Gulf regions are in Qatar and Saudi Arabia, not the Emirates.
A GPU-cloud provider, founded well before the current AI boom, that has stayed focused on one thing: renting NVIDIA GPU capacity by the hour or in reserved clusters, without a broader general-purpose cloud around it.
An "AI-native cloud" built specifically for open-source models: serverless, pay-per-token inference on 200+ models, plus dedicated GPU clusters for teams that want to train or fine-tune their own.
GPU
10 providers at this layer
The specialised chips - mostly made by NVIDIA - that do the actual arithmetic behind AI, built to run millions of small calculations in parallel far faster than a general-purpose computer chip can.
The technical version
Graphics processing units and purpose-built AI accelerators (GPUs, LPUs) whose parallel architecture is what makes training and running deep-learning models computationally feasible at scale.
Amazon Web Services (AWS)
In the UAEThe largest cloud provider by market share, and the infrastructure most AI teams still default to first - GPU instances, managed model access through Bedrock, and the storage and networking underneath both.
G42 / Core42
UAE-basedAbu Dhabi's flagship AI technology group. Core42, its cloud and AI-infrastructure company, runs sovereign AI cloud services on NVIDIA GPUs; G42 co-developed the Condor Galaxy AI supercomputer network with Cerebras. The closest thing the UAE has to its own hyperscaler.
Microsoft Azure
In the UAEMicrosoft's cloud platform, and the first of the major clouds to open a dedicated Gulf region. Azure OpenAI Service is the most common route enterprises use to buy OpenAI's models under their own compliance terms.
NVIDIA
Not in the UAEThe company whose chips sit under almost all of this stack. NVIDIA does not operate UAE infrastructure itself, but its GPUs are what every cloud and sovereign-AI provider in this catalog is actually renting out.
Oracle Cloud Infrastructure (OCI)
In the UAEOracle's cloud platform, and the operator of two UAE regions - Dubai and Abu Dhabi - the latter now home to the Middle East's first OCI Supercluster built on NVIDIA Blackwell GPUs for sovereign AI workloads.
A GPU-specialist cloud that grew out of crypto mining infrastructure into one of the largest independent NVIDIA GPU providers, built specifically for AI training and inference rather than general-purpose computing.
Google Cloud
Not in the UAEGoogle's cloud platform and home of the Gemini model family through Vertex AI. Notably absent from the UAE specifically: its nearest Gulf regions are in Qatar and Saudi Arabia, not the Emirates.
Groq
Not in the UAEGroq builds LPUs - chips designed specifically for inference rather than training - and sells access to them through GroqCloud. Its major Gulf build-out is in Saudi Arabia, not the UAE.
A GPU-cloud provider, founded well before the current AI boom, that has stayed focused on one thing: renting NVIDIA GPU capacity by the hour or in reserved clusters, without a broader general-purpose cloud around it.
An "AI-native cloud" built specifically for open-source models: serverless, pay-per-token inference on 200+ models, plus dedicated GPU clusters for teams that want to train or fine-tune their own.
Data Centre
12 providers at this layer
The physical building - racks, servers, cooling - where everything above this actually lives. An AI model has no existence outside a room full of running machines.
The technical version
The physical facility layer: colocation and hyperscale data centres, their power density and cooling architecture, and the interconnection between them - increasingly what limits how much AI compute a company or country can actually deploy.
Amazon Web Services (AWS)
In the UAEThe largest cloud provider by market share, and the infrastructure most AI teams still default to first - GPU instances, managed model access through Bedrock, and the storage and networking underneath both.
e& enterprise
UAE-basedThe enterprise technology arm of e& (formerly Etisalat), and the operator behind OneCloud - a sovereign hyperscale cloud built with Oracle, hosted and governed entirely inside the UAE.
G42 / Core42
UAE-basedAbu Dhabi's flagship AI technology group. Core42, its cloud and AI-infrastructure company, runs sovereign AI cloud services on NVIDIA GPUs; G42 co-developed the Condor Galaxy AI supercomputer network with Cerebras. The closest thing the UAE has to its own hyperscaler.
Khazna Data Centers
UAE-basedReported as the UAE's largest data-centre platform, built from the 2022 merger of G42's and e&'s domestic data-centre assets. The physical real estate underneath much of the country's sovereign cloud ambitions.
Microsoft Azure
In the UAEMicrosoft's cloud platform, and the first of the major clouds to open a dedicated Gulf region. Azure OpenAI Service is the most common route enterprises use to buy OpenAI's models under their own compliance terms.
Moro Hub (Digital DEWA)
UAE-basedThe digital arm of Dubai's electricity and water utility, and proof that in the Gulf, power infrastructure and AI infrastructure are often the same company. Its Green Data Centre runs entirely on solar power from the Mohammed bin Rashid Al Maktoum Solar Park.
Oracle Cloud Infrastructure (OCI)
In the UAEOracle's cloud platform, and the operator of two UAE regions - Dubai and Abu Dhabi - the latter now home to the Middle East's first OCI Supercluster built on NVIDIA Blackwell GPUs for sovereign AI workloads.
A GPU-specialist cloud that grew out of crypto mining infrastructure into one of the largest independent NVIDIA GPU providers, built specifically for AI training and inference rather than general-purpose computing.
du / Datamena (EITC)
UAE-basedThe data-centre and connectivity business of Emirates Integrated Telecommunications Company (du). Datamena is a carrier-neutral hub connecting UAE customers directly to major cloud providers, alongside du's own Abu Dhabi and Dubai facilities.
Equinix
In the UAEThe world's largest independent colocation and interconnection operator - not a cloud provider itself, but the neutral ground where clouds, carriers and enterprises physically connect to each other, including at its Dubai campus.
Google Cloud
Not in the UAEGoogle's cloud platform and home of the Gemini model family through Vertex AI. Notably absent from the UAE specifically: its nearest Gulf regions are in Qatar and Saudi Arabia, not the Emirates.
A GPU-cloud provider, founded well before the current AI boom, that has stayed focused on one thing: renting NVIDIA GPU capacity by the hour or in reserved clusters, without a broader general-purpose cloud around it.
Power & Network
6 providers at this layer
Everything above this needs two things nobody thinks about until they run out: electricity to run and cool the machines, and network links fast enough to move data between them.
The technical version
Power generation and grid capacity feeding data centres, plus the fibre, interconnection and edge network moving data between facilities and out to users - increasingly the binding constraint on AI build-out, not chip supply.
e& enterprise
UAE-basedThe enterprise technology arm of e& (formerly Etisalat), and the operator behind OneCloud - a sovereign hyperscale cloud built with Oracle, hosted and governed entirely inside the UAE.
Khazna Data Centers
UAE-basedReported as the UAE's largest data-centre platform, built from the 2022 merger of G42's and e&'s domestic data-centre assets. The physical real estate underneath much of the country's sovereign cloud ambitions.
Moro Hub (Digital DEWA)
UAE-basedThe digital arm of Dubai's electricity and water utility, and proof that in the Gulf, power infrastructure and AI infrastructure are often the same company. Its Green Data Centre runs entirely on solar power from the Mohammed bin Rashid Al Maktoum Solar Park.
Cloudflare
In the UAEBest known as a network and security company, Cloudflare also runs GPU inference at the edge through Workers AI - and has operated a Dubai point of presence since 2018, well before most AI-specific providers reached the region.
du / Datamena (EITC)
UAE-basedThe data-centre and connectivity business of Emirates Integrated Telecommunications Company (du). Datamena is a carrier-neutral hub connecting UAE customers directly to major cloud providers, alongside du's own Abu Dhabi and Dubai facilities.
Equinix
In the UAEThe world's largest independent colocation and interconnection operator - not a cloud provider itself, but the neutral ground where clouds, carriers and enterprises physically connect to each other, including at its Dubai campus.
What is physically in the UAE
10 of the 17 providers on this page have a documented site in the country.
Based in the UAE
- G42 / Core42
United Arab Emirates (Abu Dhabi)
- Khazna Data Centers
United Arab Emirates - described as the country's largest data-centre platform
- Moro Hub (Digital DEWA)
United Arab Emirates (Dubai)
- e& enterprise
United Arab Emirates - 11 data centres across Abu Dhabi, Dubai and Al Ain
- du / Datamena (EITC)
United Arab Emirates (Dubai, Abu Dhabi)
Global providers with UAE sites
- Amazon Web Services (AWS)
Middle East (UAE) - me-central-1, Dubai
- Microsoft Azure
UAE North - Dubai; UAE Central - Abu Dhabi
- Oracle Cloud Infrastructure (OCI)
UAE - Dubai (2020) and Abu Dhabi, me-abudhabi-1 (2021)
- Cloudflare
Middle East - Dubai, Riyadh, Doha, Kuwait City, Muscat and more
- Equinix
UAE - Dubai (DX1, DX2, DX3) and Abu Dhabi
Not in the UAE
- Google Cloud
No dedicated Google Cloud Region in the UAE as of this record's last check.
- NVIDIA
NVIDIA does not operate UAE infrastructure directly.
- Groq
Groq's Gulf build-out is in Saudi Arabia, not the UAE.
No UAE site documented: CoreWeave, Lambda, Hugging Face, Together AI.
The whole stack as a table
| Layer | What it is | Providers |
|---|---|---|
| User | Where everything starts: someone typing a question, using a search box, or opening an app that happens to have AI built into it. Nothing below this layer matters unless it eventually reaches a real person doing something with it. | Covered elsewhere |
| Application | The app, website or feature that wraps an AI model into something usable - a chatbot, a copilot inside other software, a search assistant. This is the layer most people actually mean when they say "AI." | Covered elsewhere |
| Model | The trained system that actually produces the answer - GPT, Gemini, Claude, Llama and others like them - plus the platforms that make models like these available for developers to build on. | 5 providers |
| Inference | The moment a trained model is actually used to produce an answer, in about the time it takes to read this sentence. Training a model happens once; inference happens every single time someone uses it. | 7 providers |
| Compute | The raw computing capacity - rented by the hour, the month or the year - that trains and runs AI models. This is what "the cloud" mostly means in practice. | 10 providers |
| GPU | The specialised chips - mostly made by NVIDIA - that do the actual arithmetic behind AI, built to run millions of small calculations in parallel far faster than a general-purpose computer chip can. | 10 providers |
| Data Centre | The physical building - racks, servers, cooling - where everything above this actually lives. An AI model has no existence outside a room full of running machines. | 12 providers |
| Power & Network | Everything above this needs two things nobody thinks about until they run out: electricity to run and cool the machines, and network links fast enough to move data between them. | 6 providers |
How this page is checked
- Every provider is a real company, checked against its own site or an industry source.
- Pricing links go only to the provider’s own pricing page.
- A UAE site is shown only where a region or facility is documented - otherwise the page says so.
- Last checked 24 September 2026.
Choosing between a global cloud and a UAE one?
Tell us what you are deploying and we will help you work through it. Infrastructure news and buying guides are there if you would rather read first.