R&D happens in every sector. Qualifying for it is a discipline.
The UAE R&D Tax Credit rewards genuine innovation — not routine work. Across sixteen innovation-intensive sectors, we identify precisely where the qualifying R&D sits and build the evidence to prove it, judged against the same five criteria the legislation adopts.
One test for R&D, across every sector.
The UAE regime defines qualifying R&D on the OECD Frascati Manual. To qualify, an activity must meet all five criteria — the lens we apply to every project, in every industry below.
Novel
Aimed at new knowledge or capability, not already available in the field.
Creative
Built on original, non-obvious concepts and hypotheses.
Uncertain
The outcome, or the route to it, cannot be known in advance.
Systematic
Planned, budgeted and documented as a deliberate investigation.
Transferable
Results can be reproduced and shared, not tied to one person.
Where the qualifying R&D sits.
The five criteria judge very different work the same way a fintech settlement engine, a desalination membrane, an offshore decarbonisation rig. Select a sector to see example projects and, more importantly, where the qualifying R&D actually lies within them.
01 AI & Agentic AI Machine learning, NLP and autonomous agents built for real-world constraints.
Machine learning, NLP and autonomous agents built for real-world constraints.
- Training a domain-specific model where existing architectures fail on Arabic-language or region-specific data.
- Building an agentic system that plans and executes multi-step tasks reliably under uncertainty.
- Novel techniques to reduce inference cost or latency below what off-the-shelf methods allow.
Not in calling an existing API or fine-tuning with standard tooling but in the experimental work to overcome a technical uncertainty: an architecture, training method or evaluation approach where the team genuinely does not know if it will work.
02 Deeptech & Frontier Technology Quantum, robotics, semiconductors, advanced materials and photonics.
Quantum, robotics, semiconductors, advanced materials and photonics.
- Building qubits, control electronics or sensors where coherence and error rates are genuinely uncertain.
- Autonomous manipulation, perception or navigation that resists established robotics methods.
- Novel semiconductor architectures, photonics or advanced materials with unproven performance.
Not in assembling proven components to a known spec — but in the experimental work at the frontier: a device, material or architecture where the team genuinely does not know if it can be made to work.
03 Fintech, Crypto & Digital Assets Payments, settlement, blockchain and digital-asset infrastructure.
Payments, settlement, blockchain and digital-asset infrastructure.
- A settlement engine achieving throughput or finality that available databases cannot support.
- Novel cryptographic or consensus mechanisms for a regulated digital-asset platform.
- Real-time fraud detection where accuracy demands new modelling, not rule tuning.
Not in wiring up a payment gateway or building standard CRUD screens but where you push a system beyond its known limits: latency, scale, security or a consensus problem with no established solution.
04 Cloud & SaaS Software Scalable platforms and enterprise software solving hard engineering problems.
Scalable platforms and enterprise software solving hard engineering problems.
- A multi-tenant architecture that must guarantee isolation and performance at a scale with no proven pattern.
- A data-sync or offline-first engine resolving conflicts that existing frameworks cannot.
- Re-engineering a core system to meet a performance target thought to be unreachable.
Not in routine configuration, integration or feature build but in the appreciable improvement to underlying technology, where a competent professional could not readily deduce how to achieve the result.
05 Life Sciences, Biotech & HealthTech Medical devices, diagnostics, digital health and therapeutics.
Medical devices, diagnostics, digital health and therapeutics.
- A diagnostic device improving sensitivity or specificity beyond current instruments.
- A digital-health algorithm validated to clinical-grade performance on new data.
- Formulation or bioprocess work to make a therapy stable, scalable or deliverable.
Not in routine clinical admin or standard trials management but in the scientific uncertainty: whether a mechanism, device or method actually works, and the systematic experimentation to resolve it.
06 Energy, Climate & Sustainability Renewables, storage, carbon reduction and sustainability technology.
Renewables, storage, carbon reduction and sustainability technology.
- Solar or storage systems engineered for extreme-heat, high-dust Gulf conditions.
- Grid-integration or battery-management methods with no off-the-shelf answer.
- New materials or processes that measurably cut emissions or energy intensity.
Not in installing proven kit to spec but where performance in real conditions is genuinely uncertain and demands prototyping, modelling and iterative test to resolve.
07 Oil & Gas & Energy Transition Upstream, downstream, decarbonisation and the digital oilfield.
Upstream, downstream, decarbonisation and the digital oilfield.
- Carbon capture, utilisation and storage (CCUS) pilots at a downstream facility.
- Enhanced-recovery or drilling techniques for challenging reservoir conditions.
- Low-carbon hydrogen production, or a digital-twin controlling a live process under new constraints.
Not in routine operations, maintenance or applying established engineering but in the technological uncertainty of adapting or inventing process technology: whether a capture method, recovery technique or hydrogen pathway performs at scale.
08 Construction & PropTech The built environment, modern methods of construction and smart buildings.
The built environment, modern methods of construction and smart buildings.
- Modular or offsite construction methods that must meet tolerances on-site for the first time.
- Low-carbon concrete or materials engineered for Gulf climate and loading.
- Digital twins and building-management systems integrating data no existing platform handles.
Not in standard design-and-build to code, nor routine BIM modelling but where a material, method or system carries genuine technical uncertainty resolved through prototyping and testing, not textbook engineering.
09 Advanced Manufacturing & Industry 4.0 Automation, robotics and next-generation production processes.
Automation, robotics and next-generation production processes.
- A production line achieving tolerance, yield or throughput no known process reaches.
- Robotics or vision systems for tasks that resist standard automation.
- New materials or additive-manufacturing methods requiring iterative process development.
Not in buying and commissioning standard equipment but in the process innovation: developing or adapting production technology where the outcome is uncertain and must be proven through systematic trials.
10 Telecoms, 5G, Satellite & Space Network infrastructure, connectivity and space systems.
Network infrastructure, connectivity and space systems.
- RF or antenna design achieving coverage or capacity beyond current designs.
- Low-latency 5G/edge architectures for a use case with no established pattern.
- Satellite payload, ground-station or link-optimisation work under hard physical limits.
Not in deploying vendor equipment to spec but where physics or scale impose an uncertainty that demands new design, modelling and experimental validation.
11 Cybersecurity Threat detection, encryption and security infrastructure.
Threat detection, encryption and security infrastructure.
- Detection methods identifying novel attack classes that signature-based tools miss.
- Applied cryptography — including post-quantum — for a real deployment constraint.
- Security architecture that must hold at a scale or threat model with no proven blueprint.
Not in configuring or deploying existing security products but in advancing the underlying technology, where defeating a threat requires genuinely uncertain, experimental work.
12 Agritech & Food Security Controlled-environment agriculture, water efficiency and food systems.
Controlled-environment agriculture, water efficiency and food systems.
- Vertical-farming or hydroponic systems engineered for yield in desert conditions.
- Water-use or desalination-linked methods that cut consumption beyond current practice.
- Crop, sensing or supply-chain technology validated where no proven solution exists.
Not in operating known growing systems but where adapting agriculture to an extreme environment poses a scientific or technological uncertainty resolved through systematic experiment.
13 Media, Gaming & Digital Platforms Game engines, streaming and immersive content technology.
Game engines, streaming and immersive content technology.
- Rendering, physics or netcode achieving performance a standard engine cannot.
- Streaming or compression methods that beat available codecs on a hard constraint.
- Real-time immersive systems (AR/VR) requiring novel technical solutions.
Not in content production, level design or asset creation but in the underlying engineering: an engine, pipeline or algorithm advanced beyond what existing technology readily allows.
14 Logistics, Supply Chain & Mobility Autonomous transport, supply-chain systems and smart logistics.
Autonomous transport, supply-chain systems and smart logistics.
- Perception, planning or control for autonomous vehicles in real UAE conditions no off-the-shelf stack handles.
- Real-time routing or fleet optimisation where scale and constraints defeat available solvers.
- Port, warehouse or cold-chain automation at throughput with no proven pattern.
Not in deploying vehicles, robots or software to a proven spec but where you push a system beyond its known limits: autonomy, scale or an optimisation problem with no established answer.
15 Water, Desalination & Environment Desalination, water reuse and environmental technology.
Desalination, water reuse and environmental technology.
- Novel membranes or processes that cut energy or fouling beyond current desalination practice.
- Treatment or reuse trains recovering water to a standard existing processes cannot reliably reach.
- Brine management, ZLD or environmental technology with no proven, cost-viable answer.
Not in installing proven kit or routine operation but in the process innovation: developing a membrane, chemistry or treatment train where performance at scale and cost is genuinely uncertain.
16 Aviation, Aerospace & MRO Aircraft systems, MRO engineering and aerospace manufacturing.
Aircraft systems, MRO engineering and aerospace manufacturing.
- Avionics, propulsion or control developments where performance or integration is genuinely uncertain.
- Novel repair or overhaul methods for damage or components with no established procedure.
- Composite or metal manufacturing meeting tolerances and certification no current method reaches.
Not in routine maintenance or manufacturing to a qualified process but where a system, repair method or process demands genuine, uncertain technical development to meet requirement.
The examples above are illustrative of the kind of work that commonly qualifies. Eligibility is always determined project by project against the five criteria and the current legislation — talk to us for an assessment of your own work.
Not sure where your R&D sits?
That is the assessment we do first — separating genuine R&D from routine work, before a single figure is claimed. Tell us what you are building and we will tell you honestly whether it qualifies.
