NextFin News - Presight CEO Thomas Pramotedham’s argument that organizations still need AI-led productivity gains even as geopolitical conflict intensifies cuts against an easy assumption in global markets: that war, trade fragmentation and policy uncertainty automatically delay technology spending. For a company built around sovereign AI contracts, the sharper question is not whether conflict is bad for growth in the abstract, but whether instability makes automation, data control and operational resilience more urgent. Presight’s own 2025 results suggest that, at least so far, the answer is yes. Revenue rose 36.9% year on year to AED 3.03 billion, EBITDA increased 23.5% to AED 785 million, international revenue jumped 130% and 93% of annual revenue came from multi-year contracts, according to the company’s 2025 annual report.
Those figures matter because they point to a specific business model. Presight is not selling discretionary consumer software or one-off AI experiments. It is positioning artificial intelligence as infrastructure for governments, utilities, financial institutions and industrial operators that need to keep systems running through political shocks, supply disruptions and tighter data-sovereignty rules. That makes Pramotedham’s latest message more than a management sound bite. It is a claim about where AI demand sits in the capital-spending stack when the world becomes less predictable.
The significance is broader than one Abu Dhabi-listed company. If conflict pushes national champions, regulated sectors and state-linked enterprises to treat AI as resilience infrastructure, spending may rotate away from open-ended experimentation and toward secure, local, contract-backed deployments. That would support companies exposed to sovereign compute, industrial automation, critical-infrastructure software and mission-critical analytics even if more speculative parts of the AI economy become harder to justify. The issue, then, is not whether conflict disappears. It is whether AI has crossed the line from optional efficiency tool to strategic operating requirement.
Presight’s annual report argues that this transition is already under way. The company said artificial intelligence is becoming infrastructure and that organizations integrating intelligence securely and at scale will be best positioned to lead the next phase of economic development. That framing aligns closely with Pramotedham’s latest insistence that productivity cannot be deferred simply because the geopolitical backdrop has worsened. In both cases, the underlying proposition is that unstable conditions raise the value of systems that cut waste, speed decisions and protect domestic control over data and operations.
The central tension, however, is whether investors should read that argument as a durable structural shift or as a temporary response to a difficult macro period. Conflict can raise urgency, but it can also freeze procurement, complicate supply chains and redirect state budgets toward defense or emergency spending. The article’s core judgment is that the push into sovereign and productivity-focused AI is structural, while the pace of deal conversion remains cyclical. That distinction matters because it separates long-run demand from quarter-to-quarter execution risk.
Why Conflict Can Accelerate AI Spending Rather Than Delay It
The first mechanism is straightforward but often misunderstood. Conflict and geopolitical fragmentation increase the cost of operational slack. When supply lines are less predictable, energy systems face greater stress, cyber risk rises and public institutions need faster decision loops, the economic value of automation goes up. AI in that context is not a luxury upgrade. It becomes a way to preserve throughput with fewer bottlenecks, fewer manual interventions and tighter visibility across complex systems.
Presight’s own business mix supports that reading. The company described its focus areas as public services, energy and utilities, finance and smart infrastructure. Those sectors do not buy technology purely for growth optionality. They buy systems that can lower failure rates, shorten response times and improve resource allocation in high-stakes environments. In energy, for example, automation tools can reduce downtime and optimize production. In finance, secure AI infrastructure can improve surveillance, compliance and system resilience. In public-sector settings, data platforms can help allocate resources faster when logistics become harder to manage. That is the transmission channel from conflict to demand: instability raises the premium on productivity and control.
"Artificial intelligence is becoming infrastructure."
That line from Presight’s 2025 annual report is more than branding. Infrastructure spending behaves differently from experimental software budgets. Once an enterprise or government defines AI as part of its operating backbone, procurement logic changes. Multi-year contracts, local deployment requirements, security audits and integration work start to dominate. That helps explain why 93% of Presight’s FY25 revenue came from multi-year contracts, according to the company. A revenue base tied to long-duration commitments is consistent with infrastructure-like demand, not pilot-project volatility.
The second mechanism is sovereignty. Conflict does not just raise the need for productivity; it changes who organizations trust to provide that capability and where data and compute are allowed to sit. Across governments and regulated industries, the more uncertain the geopolitical environment becomes, the stronger the incentive to localize sensitive workloads, control data flows and reduce dependence on external technology stacks that may become politically or commercially constrained. Presight has made sovereign AI a central part of its proposition, and that positioning becomes more valuable when cross-border technology access is no longer assumed to be frictionless.
This is where the company’s international growth matters. International revenue grew 130% in 2025 and represented 38.5% of total revenue, while new orders reached AED 3.4 billion, according to the annual report. Those are not trivial figures for a company still relatively early in its life as a listed entity. They suggest that the sovereign AI pitch is translating beyond its domestic base. The business implication is that geopolitical fragmentation may enlarge the addressable market for trusted national or regional AI partners rather than shrink it outright.
That is also the point at which the market may be underestimating second-order effects. The obvious first-order conclusion is that conflict hurts growth and makes boards more cautious. The second-order conclusion is more selective: it can make spending harder to justify for discretionary projects while making it easier to justify for efficiency tools tied to resilience, compliance and national control. In other words, instability can compress the market for general-purpose experimentation while deepening demand for mission-critical deployment. That is a narrower but more durable pool of AI spending.
Look at the sectors Presight highlights. Energy systems, financial infrastructure and government operations are all areas where the cost of delayed decisions rises rapidly during periods of tension. In such settings, AI is not just about replacing labor or cutting back-office expense. It is about compressing reaction times, improving pattern recognition and maintaining service continuity. When management teams say productivity matters despite conflict, the market should read productivity in that broader operating sense, not simply as a euphemism for headcount reduction.
There is another important channel. Conflict often sharpens political pressure for domestic economic performance. Governments facing external risk need public services to work, logistics to stay efficient, utilities to remain stable and industrial output to avoid unnecessary losses. That can make AI deployment politically easier in exactly those use cases where measurable performance gains can be shown. Presight’s February 2025 launch of its Presight Synergy platform framed enterprise AI adoption around solving complexity and inefficiency, and that language fits the current environment well: buyers are more likely to fund projects with visible operational returns than open-ended innovation agendas.
The implication is not that every AI company wins from conflict. Many do not. Businesses dependent on globally smooth hardware supply, cross-border data mobility, abundant venture funding or lightly governed experimentation could still face pressure. The point is that Presight’s niche sits closer to protected budgets than many of the market’s better-known AI trades. That distinction becomes sharper when macro uncertainty rises.
Structural Demand, Cyclical Conversion
The cleanest way to understand the story is to separate the structural leg from the cyclical leg. Structurally, the evidence points to a lasting shift toward sovereign, secure and productivity-focused AI. Cyclically, quarterly revenue recognition, procurement timing and expansion margins can still swing with politics, funding cycles and implementation bottlenecks. Blending those two forces would blur the demand case and the risk case at the same time.
The structural argument rests on more than one year of growth. Presight said it has delivered 12 consecutive quarters of growth since its IPO. It also said backlog increased year on year despite strong revenue conversion. Those two details matter because structural demand should show up not only in current-period revenue but also in the persistence of contracted work and the ability to replenish future business while scaling current delivery. The company’s debt-free balance sheet adds another support point: it suggests the model has so far been able to expand without depending on fragile external financing conditions.
Evidence of structural demand also appears in the type of customer problem Presight is addressing. The company’s annual report repeatedly describes AI as sovereign, mission-critical and embedded in economic architecture. That language implies a regime change in how buyers classify AI expenditure. If AI sits inside budget lines for public safety, energy optimization, financial infrastructure and national digital systems, demand behaves less like cyclical software experimentation and more like long-horizon capability building. History matters here. Previous enterprise technology booms often produced short-lived pilot programs that failed to scale once macro conditions tightened. The current sovereign AI build-out looks different because the purchasing logic is tied to state capacity, regulatory control and infrastructure resilience, not just office productivity or corporate innovation budgets.
Still, the cyclical element should not be minimized. Multi-year contracts can protect revenue visibility, but they do not eliminate the timing risk around awards, deployments, milestone payments or margin capture. Conflict can slow approvals, alter funding priorities or create friction in sourcing critical components. Even a structural build-out can move in bursts rather than a straight line. That is why the right conclusion is not that Presight is immune to macro stress. It is that the company appears exposed to a form of AI demand with a deeper floor than the market might assume.
The historical comparison also matters. In past technology downcycles, spending tied directly to measurable efficiency often held up better than spending tied to exploratory transformation programs. Buyers cut pilots first and preserve systems that can lower operating cost, improve security or sustain output. Presight’s emphasis on platforms for utilities, finance, government and industrial users places it in the part of the stack that organizations are less willing to suspend. That does not make near-term execution effortless, but it does support the idea that the core demand signal is mean-resistant rather than purely sentiment-driven.
This is where the cyclical-versus-structural test becomes useful. A cyclical story would require evidence that demand is mainly a temporary reaction to unusually high uncertainty and that it should revert once the external shock fades. A structural story requires evidence that institutions have changed how they define the problem itself. Presight’s annual report provides support for the second view: it frames AI as national and enterprise infrastructure, highlights long-duration contracts and shows growth spreading internationally as sovereign adoption broadens. That is not proof of permanence, but it is stronger than a one-off sales boom.
The second-order question is whether this demand is already fully priced into the broader AI narrative. Probably not in this specific form. Public markets have heavily rewarded the upstream beneficiaries of AI, particularly compute, semiconductors and hyperscale infrastructure. But the monetization layer tied to sovereign deployment, regulated-sector productivity and national capability building is less uniform and less universally understood. If markets continue to focus primarily on model capacity and chip access, they may underprice the companies that turn those capabilities into embedded operating systems for governments and critical industries.
That could matter for how future results are interpreted. A period of slower deal conversion but stable backlog growth could be read negatively in a generic software framework, yet more neutrally in an infrastructure framework where contracts are larger, more complex and more politically conditioned. Conversely, a sharp increase in international orders or backlog quality may say more about the structural demand environment than a short-term margin move. The right lens changes what counts as signal.
"Organizations that integrate intelligence securely and at scale will be best positioned to lead the next phase of economic development."
That statement from Presight’s annual report crystallizes the structural thesis. The company is arguing that AI adoption is moving from advantage-seeking to capability-preserving. In calmer macro periods, boards can treat digital transformation as a route to better performance. In more fractured periods, they can treat it as insurance against operational decay. The former can be postponed. The latter is harder to ignore.
There is a useful analogy here, used carefully: conflict can act like a stress test on an operating model. Systems that looked adequate in benign conditions often reveal hidden inefficiencies when the environment becomes volatile. AI platforms designed to reduce those inefficiencies may therefore see stronger demand precisely when headline risk is rising. That does not reverse the macro drag from conflict. It changes which budgets become hardest to cut.
The Counter-Thesis: Conflict Can Freeze Ambition Before It Funds Automation
The strongest case against the structural thesis is serious and should not be brushed aside. Conflict can drain public budgets, weaken business confidence and disrupt technology supply chains. It can also make procurement more political and stretch project timelines even where strategic intent remains intact. Under that view, executives may talk more about resilience and productivity precisely because actual spending becomes harder to execute. The rhetoric strengthens while conversion slows.
This counter-thesis attacks the argument at its foundation. If geopolitical stress mainly delays implementation, then sovereign AI could remain an attractive story without becoming a reliably monetizable one. Large public-sector contracts can be lumpy in any environment. Add export controls, shifting alliances, budget reallocations and slower approvals, and the gap between strategic necessity and recognized revenue can widen materially. For a company like Presight, that would mean caution remains justified even if management language sounds compelling.
There are reasons to take that challenge seriously. The same geopolitical conditions that make data sovereignty more valuable can also complicate hardware access, raise local-content requirements and slow integration with global partners. A company can win mandates in principle while facing delays in execution. And because sovereign AI projects often involve multiple agencies, regulated workflows and high security requirements, they can absorb macro friction more readily than ordinary enterprise software deals.
Even Presight’s positive data should be read with care. Revenue growth of 36.9%, EBITDA growth of 23.5% and international revenue growth of 130% are backward-looking 2025 figures. They show momentum, not immunity. The right question is whether those metrics remain durable as the geopolitical environment becomes more contested and as buyers move from enthusiasm to operational delivery. Structural demand is only fully credible if the backlog, renewal profile and international diversification continue to hold when financing, logistics and policy coordination get harder.
The answer to the counter-thesis is not that it is wrong in every respect. It is that it likely overstates the symmetry between discretionary and mission-critical AI spending. If Presight were selling generalized experimentation, the bear case would be stronger. But the company’s concentration in government-linked and critical-industry use cases means the spending hurdle is different. Budgets can shift, but the need to protect energy systems, financial infrastructure and national data control does not disappear when conflict rises. In some cases, that is exactly when such budgets gain priority.
The more nuanced conclusion is that conflict changes the spending mix before it changes the spending total. Boards and governments may defer broad digital ambitions and tighten return thresholds, while still accelerating projects tied to measurable productivity, resilience and sovereignty. That means the path can become narrower and more selective without becoming smaller. Companies that cannot prove hard operational value may lose out. Companies that can may gain share of a more disciplined pool.
The single most important falsifying signal for the structural thesis is straightforward: if Presight’s backlog and the share of revenue coming from multi-year contracts begin to decline meaningfully over successive reporting periods while sovereign AI demand rhetoric remains elevated, then the thesis that AI has moved into protected, infrastructure-like budget lines would weaken sharply. Put differently, the story fails if strategic urgency does not continue to translate into durable contracted work.
That is the threshold the market should watch more closely than broad commentary about AI excitement. Words are cheap. Contract structure is not.
What This Means for AI Markets and the Road Ahead
The broader implication of Pramotedham’s message is that the AI economy may be entering a more differentiated phase. The first phase rewarded access to compute, models and capital. The next phase may reward companies that can embed AI into national and industrial operating systems under tighter political, regulatory and security constraints. That shift does not replace the first phase, but it does change where incremental resilience in demand may come from.
In the short term, sentiment can still swing with headlines about war, trade restrictions, export approvals and government budgets. That means valuations for AI-linked companies may remain volatile even where the underlying use case is strengthening. For Presight specifically, short-term perception is likely to depend on contract announcements, backlog conversion and signs that international expansion continues beyond the UAE base. Readers looking only at headline macro risk may miss that tension: bad news for global confidence can coexist with firmer demand for operational AI.
Over the medium term, fundamentals should hinge on whether Presight can keep proving that sovereign AI is not just a policy slogan but a repeatable contract model. The 2025 figures provide a strong starting point: AED 3.03 billion of revenue, AED 785 million of EBITDA, AED 3.4 billion of new orders, 38.5% of revenue from international markets and 93% of revenue tied to multi-year contracts. The next test is whether those markers keep compounding as projects scale in more jurisdictions and across more regulated sectors.
Over the long term, the structural thesis depends on whether AI remains classified as infrastructure rather than reverting to a more discretionary software bucket. If governments and national enterprises continue to embed AI into energy, finance, public safety and industrial systems, the demand base should prove more durable than the typical enterprise-tech cycle. If, however, adoption stalls at pilot level or sovereign rhetoric fails to sustain contract depth, the category will look more cyclical than today’s champions suggest.
The base case is that conflict and fragmentation continue to favor a narrower class of AI providers whose products can show measurable productivity gains and satisfy sovereignty requirements. The upside case is that more countries accelerate national AI programs, expanding the market for trusted deployment partners faster than current expectations imply. The downside case is that procurement slows materially or budget redirection delays revenue conversion despite healthy strategic intent. Each scenario turns on contract durability, not on abstract enthusiasm for AI.
That is why the next catalysts matter. Future results should be judged through backlog quality, international order growth, the share of recurring or multi-year revenue, and evidence that Presight’s platforms continue to move from concept to embedded workflow in finance, utilities, government and industrial settings. Those are the operating signals that will show whether Pramotedham’s claim about productivity demand surviving conflict is merely rhetorically plausible or economically true. As of Aug. 12, 2026, the clearest evidence available remains the company’s 2025 annual report and publicly released product positioning.
The market’s first instinct is to treat conflict as a universal tax on technology spending. That is too blunt. For sovereign and productivity-focused AI, instability may be less a stop sign than a sorting mechanism.
That is the sharper read of Presight’s message: conflict does not erase the need for AI productivity; it reveals which forms of AI spending were optional all along.
Explore more exclusive insights at nextfin.ai.
