Intelligence Is the New Infrastructure: The Middle East’s DeepTech Decade with 2035 Predictions

The Middle East is no longer importing the future. It is designing one. Dr. Ankur Narang, Founder of DeepCoreX Labs, on AI, quantum, and the venture studio decade ahead.

I. The Room With No Software in It

Three months ago, in a room overlooking the Abu Dhabi Corniche, a sovereign wealth fund executive set down his coffee and asked the question that has come to define this decade. “How do we stop renting our future?” He was not talking about real estate.

For the previous hour, we had moved through the standard agenda of such meetings: model partnerships, compute allocations, talent pipelines, regulatory architecture. Two years ago, the same conversation would have ended with a procurement decision. Which vendor. Which licence. Which integrator. That is no longer the conversation in the Gulf. The vocabulary has shifted from buying to building, from deploying to owning, from adopting to architecting. The room contained no discussion of software at all. This is the most underreported macroeconomic story of our time.

For seventy years, the global influence of the Gulf Cooperation Council was measured in barrels per day. The capital that flowed from that influence built ports, airlines, sovereign portfolios, and some of the most ambitious cities in human history. But it remained, fundamentally, an economy denominated in extracted molecules. What is happening now is something categorically different. It is the deliberate, capitalised, state-orchestrated reconstruction of the regional economy around a new substrate altogether: intelligence itself.

The scale is no longer theoretical. UAE’s MGX is racing toward more than $100 billion in assets under management, backed by sovereign wealth and aiming to deploy up to $10 billion in deals annually. Saudi Arabia’s Public Investment Fund launched HUMAIN in May 2025 to develop AI infrastructure, models and applications on top of a sovereign asset base exceeding $930 billion. In 2024, PIF spearheaded the $100 billion Project Transcendence initiative. In 2025, Qatar’s QIA launched its own AI vehicle, Qai. Total committed capital from Gulf sovereign wealth funds to the global technology stack reached roughly $2.5 trillion by early 2026.

This is not a region buying into the future. This is a region designing one. And the question my colleague asked over coffee is the right one, because the next decade will not be defined by which countries spent the most on artificial intelligence. It will be defined by which countries learned to produce it.

“The conversation in the Gulf has moved from buying to building, from deploying to owning. The room I sat in contained no discussion of software at all.”

II. Why the Gulf Is Structurally Built for DeepTech Leadership

Deep technology is a different game from consumer software, and the failure to grasp this distinction is why most analyses of the Gulf’s AI ambitions miss the point. Consumer software rewards velocity, distribution, and brand. DeepTech rewards something else entirely: patient capital, executive coherence, and the willingness to build infrastructure ahead of demand. The Gulf possesses all three at a scale almost no other region can match.

Patient capital first. A frontier foundation model can take three years to train. A quantum coherence platform can take five years to stabilise. A vertically integrated robotics company can take a decade to scale. Most venture ecosystems cannot tolerate that timeline. Sovereign capital can. When MGX co-finances next-generation AI campuses requiring deep capital and long-term commitments, with some supported facilities exceeding one gigawatt of power draw and ranking among the largest AI facilities ever planned, that is not venture capital behaviour. That is industrial policy executed at venture speed.

Executive coherence second. This is the variable Western analysts most consistently miss. In most jurisdictions, the gap between a national strategy and its execution is measured in administrations. In the Gulf, that gap is measured in quarters. When a Sheikh, a Crown Prince, a sovereign fund chair and a regulator can sit in the same room and reach a binding decision, frictional time collapses. The result is what you now see: UAE partnerships with Microsoft and OpenAI, the $10 billion Google–HUMAIN deal, the $20 billion Qai–Brookfield AI joint venture. None of it is accidental, and all of it is choreographed at a speed Western democracies structurally cannot match.

Building infrastructure ahead of demand is the third structural advantage. Critics of the Gulf’s AI buildout describe it as overcapacity. They are using the wrong framework. Saudi Arabia’s Data and AI Authority awarded a $2.7 billion contract for the 480 MW Hexagon data centre in Riyadh in January 2026, with national ambitions to develop between three and six gigawatts of AI computing capacity. Compared against current regional demand, this looks like overbuilding. Compared against the deliberate Gulf playbook that produced Dubai’s airport before there was anyone to fly there, and Aramco’s downstream petrochemicals before there was a regional customer base, it is exactly on script. Infrastructure here is a magnet, not a response.

Three structural assets compound these advantages further. Energy is the first. Training trillion-parameter models and running continuous inference at scale require gigawatt-grade power that the Western grid cannot easily provision. The Gulf has the legacy energy base and aggressive renewable transitions to underwrite this. The International Energy Agency estimates global cumulative data centre investment of $3.9 trillion between 2026 and 2030, a figure too large to be funded from the balance sheets of AI companies alone. Sovereign energy plus sovereign capital is not a luxury in that equation. It is a prerequisite.

The second is regulatory agility. The UAE released its National AI Strategy in 2017, earlier than most G20 economies, and has since institutionalised it through bodies that can move at commercial speed. When the Technology Innovation Institute released Falcon as open weights, it proved that a state-organised research stack could produce frontier-grade artefacts on a Western timeline. The third is geography and demography. The GCC sits within four hours of three billion people. Its workforce is young, multilingual, and increasingly technical. Mohamed bin Zayed University of Artificial Intelligence is the first graduate-level AI university in the world, and its trajectory illustrates a deeper national intent: build the talent locally, not borrow it permanently.

What this combination produces is something none of the traditional technology capitals can replicate: an environment in which deep technology can be built, capitalised, regulated and deployed inside a single decision-making perimeter. That is the Gulf’s structural edge. The strategic question is no longer whether the region will lead in DeepTech. It is which institutions will translate that structural edge into companies.

“Sovereign capital and gigawatt infrastructure are not enough. The Gulf’s edge becomes a Gulf advantage only when it is matched by institutions that can convert science into companies.”

III. Intelligence Is the New Infrastructure

The single most consequential intellectual move any policymaker can make in 2026 is to stop thinking about AI as software and start thinking about it as infrastructure. Software is procured. Infrastructure is built. Software is a line item. Infrastructure is sovereignty.

This reframing changes three things about how a country must behave. First, compute becomes a strategic reserve, indistinguishable in importance from oil storage, grain reserves or fibre backbones. Nations that can host frontier-scale training runs domestically will dictate the terms on which other nations access them. Nations that cannot will pay the rent indefinitely. The decision to build six gigawatts of AI compute in Saudi Arabia and to underwrite multi-billion-dollar campuses in the UAE is not a vanity expenditure. It is the equivalent of building the Aramco refining base in the 1970s: infrastructure that compounds for half a century.

Second, foundation models become institutions. A model trained on a nation’s language, regulation, history, and economic data is not a product. It is cognitive sovereignty. This is why the UAE invested in Falcon and MBZUAI’s research stack, why Saudi Arabia is building Arabic-native models through HUMAIN, and why every serious country in the next five years will field at least one sovereign foundation model. The countries that do will keep their futures. The countries that do not will rent them.

Third, and this is the part most leaders still underestimate, agentic AI rewrites the operating system of the enterprise itself. Conventional AI tools are bolted onto existing workflows. Agentic AI replaces the workflows. A bank that automates onboarding does not save thirty percent on a process. Over five years, it disassembles and rebuilds the process from scratch. The same is true for healthcare claim adjudication, sovereign procurement, port logistics, energy trading, sovereign wealth allocation, and clinical decision support.

In the Gulf, where state-owned enterprises and large family conglomerates form the spine of the economy, the economic impact of agentic AI will be larger and faster than in almost any Western market. The regulatory clearance is easier. The executive sponsorship is higher. The appetite for systemic redesign is greater. A CHRO at a Gulf national champion can authorise the redesign of an entire HR function in a single quarter. Her counterpart in a listed Western enterprise will spend that quarter briefing the board on whether to commission a pilot.

This is the unstated reason the Gulf will lead in applied AI before it leads in foundational AI. Other regions have larger research bases. None has a faster, more decisive path from pilot to production. This is not a future state. It is happening now. When Anthropic raised its latest funding round, MGX joined as a co-lead investor. Both the Qatar Investment Authority and MGX participated in xAI’s $20 billion round. Gulf capital is not following the AI revolution. It is increasingly underwriting it.

“Sovereign foundation models are not products. They are institutions. The nations that build their own will keep their futures. The rest will rent them.”

IV. Quantum Computing: The Next Strategic Enclosure

If artificial intelligence is the defining infrastructure of this decade, quantum computing is the defining infrastructure of the next. The mistake most boards and most ministries make is to treat quantum as a distant science project. It is not. It is happening on a timeline that matches, almost exactly, the planning horizons of the sovereign funds and national champions now committing to it.

The numbers under-describe the consequence. The global quantum computing market is projected to grow from $3.52 billion in 2025 to $20.20 billion by 2030, at a compound annual growth rate of 41.8 percent. McKinsey projects the total quantum technology market reaching $198 billion by 2040, with quantum computing alone capturing $28 to $72 billion by 2035. But the financial sizing is a lagging indicator. Quantum is not merely a faster classical computer. It is a different machine, suited to a specific class of problems, including molecular simulation, optimisation, cryptography and materials design, that classical machines cannot solve at relevant scales no matter how large they grow.

The first nation to industrialise quantum advantage in any of these domains will hold an asymmetric edge for a generation. A critical strategic error emerging ecosystems make is competing for hardware dominance. The race to build superconducting qubits and dilution refrigerators is capital-intensive, geographically concentrated, and increasingly subject to export controls. The smarter play, and the one we have built the DeepCoreX Quantum Center of Excellence around, is to dominate the algorithmic, applied and workforce layers, where economic value will accrue regardless of which hardware platform eventually wins.

Three quantum sub-fields deserve specific Gulf attention. Quantum computing applications are within five to ten years of delivering verifiable commercial advantage in chemistry, optimisation and machine learning. The implications for hydrocarbons, petrochemicals, pharmaceuticals, logistics and finance are enormous and disproportionately consequential for the Gulf. A hybrid quantum-classical algorithm optimising container routing at Jebel Ali compounds into billions in efficiency over a decade. A quantum chemistry workflow applied to Aramco’s downstream catalysis or to Saudi pharmaceutical molecule discovery produces results that no classical platform can match.

Quantum sensing is the use of quantum effects to measure gravity, magnetic fields, time and inertia at precisions previously impossible. Applications run from sub-surface oil, mineral and water exploration to GPS-denied navigation, advanced medical imaging and defence-grade situational awareness. Quantum communication and post-quantum cryptography are the layers that will eventually carry traffic immune to both classical and quantum interception. These technologies are forecast to become critical to governments whose pipelines, ports, payment rails and sovereign databases will need post-quantum security within the decade.

The binding constraint across all three is talent. Global quantum workforce demand is projected to outstrip supply by roughly two to one, with demand approaching 10,000 qualified workers against a global pool of under 5,000. No country can buy its way out of that gap. It can only educate its way out, which is the explicit mandate of the DeepCoreX Academy and the QCoE working together. A nation cannot lead in quantum without a generation of locally trained quantum engineers. The Gulf can train them faster than almost anywhere else if it begins now.

“Quantum is not a faster classical computer. It is a different machine for problems the world cannot otherwise solve.”

V. The Venture Studio: The Missing Institution

The greatest failure of the global innovation system is not a shortage of ideas. It is the chasm between scientific breakthrough and commercial product. Most regions of the world have built either the input side of the equation, meaning universities, research labs and sovereign capital, or the output side: stock exchanges, large enterprise customers, distribution networks. Almost no region has built the connective tissue between them. That is the institution we now urgently need: the DeepTech venture studio.

The traditional venture capital ecosystem, refined over four decades for software-as-a-service and consumer internet, is structurally ill-equipped for the realities of DeepTech. When a standard venture firm evaluates a consumer application, the underlying technology is commoditised, and the risk lives in distribution. DeepTech inverts this. The distribution risk is small. There is no shortage of demand for a quantum-optimised supply chain or an agentic claims adjudication system. The risk lives in the science itself.

An algorithmic breakthrough in autonomous robotics or quantum machine learning requires multi-year R&D cycles, capital-intensive compute, and deep enterprise integration long before a first dollar of revenue. Traditional venture firms wait on the sidelines until that technical risk has been eliminated. They are good at catching falling stars. They are constitutionally incapable of building the rockets in the first place.

This is the structural void the venture studio is built to fill. A venture studio is not an accelerator. It is not an incubator. It is not a venture firm. It is a permanent operating team, made up of engineers, scientists, designers, GTM strategists, lawyers and operators, that co-creates ventures from raw scientific insight, deploys them with anchor customers, capitalises them with patient capital, and stays involved through scale. It is a company-building factory, not a portfolio aggregator.

The studio model collapses the so-called “valley of death” for deep technology. By the time a company emerges from a serious DeepTech studio, it has already achieved enterprise validation, contracted pilot revenue, and proof of algorithmic efficacy. The risk that traditional venture capital had to underwrite at seed has already been priced out. What the studio offers institutional capital is not a speculative bet on a slide deck. It is a de-risked, productised, validated company looking for growth equity.

This is the model the Gulf needs at scale. Gulf sovereign wealth funds have already built infrastructure and acquired stakes in foreign AI champions, but the regional capacity to originate venture-scale DeepTech companies remains thin. A region of this capital density and ambition should have five to ten serious DeepTech studios operating between Abu Dhabi, Dubai, Riyadh, Doha and Jeddah by 2030. There are currently far fewer. The vacuum is not a market failure. It is a market opening.

“The traditional venture firm catches falling stars. The venture studio builds the rocket. DeepTech requires the second institution, not the first.”

VI. The DeepCoreX Model: A Circulatory System, Not a Portfolio

DeepCoreX Labs was built for this moment. We operate as four interlocking units that together represent the full DeepTech value chain. But the four are not parallel businesses. They are a single circulatory system.

The Venture Studio is where companies are created. We invert the standard venture flow. Rather than waiting for founders to pitch us polished business plans, we start with high-value, high-defensibility problems, including sovereign-grade inefficiencies in energy grid optimisation, autonomous logistics, agentic enterprise workflows, and applied quantum optimisation. We then assemble the founding team, intellectual property, anchor customers and capital structure around them.

The AI Solutions Lab is where revenue and proof live. The Lab runs applied R&D and enterprise pilots, the work of taking frontier machine learning techniques and turning them into deployable production systems for banks, hospitals, energy majors, sovereign departments, ports and logistics platforms. The Lab generates the revenue that funds the studio’s longer-horizon risk capital while simultaneously producing the deployment evidence that converts research into investable companies.

DeepCoreX Academy is where the workforce is built. We train engineers, technologists and senior executives in agentic AI, applied machine learning, quantum computing fundamentals, and robotics through intensive cohort programmes, university partnerships, and corporate upskilling tracks. The Academy is not a marketing function. It is the talent engine without which the rest of the system cannot function.

The Quantum Center of Excellence (QCoE) is where the long-horizon bet lives. The QCoE runs applied research, university partnerships and training across quantum computing, quantum sensing, and quantum communication. It is designed to bypass the theoretical delays that have historically kept quantum confined to academia, and to deliver applied algorithmic IP that the Gulf can deploy in production environments inside this decade, not the next one.

The four units circulate continuously. The Academy trains people who staff the Lab. The Lab pilots technology that becomes Studio ventures. The Studio creates companies that need quantum capabilities, which the QCoE develops. The QCoE produces research the Academy then teaches. Each unit makes every other unit more valuable. We are not a portfolio of services. We are a single, integrated commercialisation engine.

“DeepCoreX is not a portfolio. It is a circulatory system: Studio, Lab, Academy and QCoE feeding one another continuously.”

VII. The India–Middle East Innovation Corridor

There is one geography that will define our work as much as the Gulf itself, and that geography is India. The India–Middle East innovation corridor is the single most underrated structural advantage in the global technology economy. It is being built right now, and the institutions that recognise it early will dominate the next decade.

India produces more STEM graduates per year than any country except China. The IITs, IISc, IIITs and a small set of elite private universities form an engineering talent pool that no Western country matches in scale. The historical constraint has not been the supply of talent. It has been the absence of an applied venture-creation layer that could translate that talent into globally significant companies.

The Gulf supplies the precise missing complement. It has the patient capital. It has the customers, namely sovereign enterprises whose digital transformation budgets exceed the GDPs of small European countries. It has the regulatory clarity. It has the geography. It has the political will.

DeepCoreX Labs is structured to operate across this corridor by design. Our model with Indian universities is straightforward and replicable. We partner with leading institutions such as IIT Delhi, IISc, and the next tier of research universities to co-establish Centres of Excellence in agentic AI, quantum computing and robotics. We contribute applied curriculum, industry projects, faculty co-development, and a route from doctoral research to deployed product.

The mechanism is concrete, not ceremonial. When an Indian doctoral candidate develops a novel agentic AI workflow at IIT Delhi, the DeepCoreX Academy creates the path to deploy that workflow in a sandboxed, anonymised pilot inside a UAE or Saudi enterprise within months, not years. The researcher gets validation against real-world data constraints they could never access in India. The Gulf enterprise gets early access to applied research it could not source domestically. DeepCoreX gets the IP, the venture seed, and the founding team.

This is not a transactional sponsorship. It is a corridor architecture. Indian researchers gain deployment markets that Bengaluru alone cannot provide. Gulf clients gain research talent that no domestic university can yet produce at the required scale. And DeepCoreX sits at the intersection, building companies that belong to both geographies and to neither, with IP, customers and capital structures spanning the corridor.

“The India–Middle East corridor is not a trade deal. It is an architecture for converting India’s largest export, engineering talent, into co-owned, globally scaled intellectual property.”

VIII. Five Predictions for 2035

1. Agentic AI Will Run Material Parts of Gulf Sovereign Wealth Operations

By 2028, multi-agent systems will execute portfolio rebalancing, deal screening, due diligence first-pass review, and risk analytics across at least three of the major Gulf sovereign vehicles. This will not eliminate human capital allocators. It will compress their decision cycles by an order of magnitude, freeing them for the contested, judgment-heavy decisions that justify their existence.

2. Quantum Advantage Will Be Proven in Logistics or Hydrocarbons

By 2030, a major Gulf port operator or a national oil company will publicly verify a hybrid quantum-classical algorithm producing double-digit improvements in routing, scheduling or molecular simulation. It will be the first commercially audited quantum advantage achieved outside the traditional US-EU-China triad.

3. Venture Studios Will Replace Traditional Early-Stage DeepTech Investing

By 2032, institutional limited partners, including pension funds, sovereign funds and family offices, will refuse to fund DeepTech at seed unless the company has been incubated, validated and de-risked inside a studio environment. The studio model will become the default operating layer for frontier technology venture creation.

4. The India–Middle East Corridor Will Outperform Europe in DeepTech Unicorn Creation

By 2033, the India–Middle East corridor will produce more DeepTech unicorns annually than any single European country. The combination of Indian engineering supply and Gulf demand-side capital is structurally larger than what any single European jurisdiction can field.

5. Gulf Exchanges Will Be Dominated by AI, Quantum and Robotics IPOs

By 2035, the largest single category of new IPOs from Gulf exchanges will not be real estate, energy or financial services. It will be applied AI, quantum and robotics. The composition of the regional stock exchanges in 2035 will look almost unrecognisable next to their 2020 composition.

These predictions will be wrong in detail. They will not be wrong in direction. And the ecosystem builders who position now will own a disproportionate share of the upside when they prove correct.

“By 2035, the largest single category of new IPOs from Gulf exchanges will not be real estate, energy or financial services. It will be applied AI, quantum and robotics.”

IX. To the Builders

There is a temptation, when writing about an era this consequential, to retreat into abstraction. I would rather be specific. The next decade will not reward those who understood the science. It will reward those who deployed it.

To sovereign funds, family offices and policymakers: the highest-return decisions you will make in this decade are not in your traded book. They are in the venture and infrastructure layer where intelligence is being industrialised. Move with operating partners who can build, not just allocate. Recognise the venture studio not as one of many fund structures but as a strategic asset.

To enterprise leaders and CTOs across the Gulf: your competitive position five years from now will be determined less by what models you adopt and more by whether you rebuild your operating system around agentic intelligence. Your proprietary data is the most valuable asset on your balance sheet. Stop renting generic AI tools and start commissioning sovereign architectures designed against your hardest operational problems.

To university vice chancellors in India and the Gulf: the institution-to-industry pipeline is the strategic asset of the next decade. The universities that treat industry collaboration as a fundraising activity will fall behind. The universities that treat it as the through-line of their relevance will define the era.

To DeepTech founders: the capital exists. The customers exist. The infrastructure exists. The remaining variable is your ambition and the institutional partners you choose. Do not attempt to navigate the early-stage friction of quantum, robotics or frontier AI alone. Choose a studio environment that gives you capital, compute, customers and validation from day one.

And to the young engineers and scientists reading this in Delhi, Bengaluru, Hyderabad, Abu Dhabi, Riyadh, Doha and everywhere in between: the question to ask yourself is not which company to join. It is which problem to dedicate the next decade of your life to solving. The infrastructure of the next century is being built right now, in real time, by people who decided not to wait.

The Gulf did not need to imitate Silicon Valley. It needed to skip an era. It is doing so. The next decade belongs to those who convert science into companies, intelligence into infrastructure, and ambition into institutions.

We intend to be among them. The room with no software in it is open. Come build.

Author Bio

Dr. Ankur Narang is a globally experienced AI and DeepTech leader with over three decades of experience in high-growth technology companies. A globally renowned scientist, he contributed to the development of first-generation supercomputers and worked alongside the team that created the Java language at Sun Microsystems in the late 1990s. He was also part of the team at IBM Research behind the Blue Gene supercomputer, which received the National Medal of Technology from the President of the USA.

Beyond research, Dr. Narang has built and led AI and DeepTech functions across a remarkably broad set of industry verticals, including Retail, EdTech, Oil & Gas, Telecom, Gaming, Messaging, Media, and Travel Technology.This cross-sectoral operating experience, spanning the world’s largest consumer platforms and the most demanding industrial environments, sits at the heart of how DeepCoreX Labs designs ventures and enterprise pilots today. He holds 15 US patents and has authored over 70 research publications.

Dr. Narang holds both a B.Tech and a Ph.D. in Computer Science from the Indian Institute of Technology Delhi, and an M.S. from Santa Clara University, placing him in a rare category of leaders who combine top-tier Indian academic rigour with global research exposure and large-scale operating experience across multiple industries.

---

Important Disclaimer

Market, regulatory, geopolitical and technological developments may cause actual outcomes to differ materially. Nothing in this article constitutes investment, legal, tax or strategic advice. References to specific companies, sovereign funds, sectors or transactions are illustrative and should not be interpreted as recommendations. Readers considering any action based on the views expressed here should consult appropriately qualified independent advisors. The views expressed are those of the author and do not necessarily reflect the views of StrongYes, its editorial board, or any affiliated entity.