Sanctions were designed as instruments of control, intended to constrain behaviour by restricting access to markets, capital, and trade. In practice, however, they have evolved into catalysts for systemic adaptation. From Iran’s shadow oil networks to Russia’s rerouted exports, modern sanctions have not halted economic activity—they have reconfigured it into parallel systems operating with reduced transparency and increased complexity. Enforcement gaps, logistical innovation, and financial engineering have allowed trade to persist beyond traditional oversight, creating a fragmented global economy where visibility diminishes as resilience increases. What was once a tool of pressure is now a mechanism of redirection, reshaping global flows rather than stopping them. This editorial examines how sanctions are no longer containing systems—but decentralising them.

Sanctions are built on a premise of control, the assumption that access to markets, financial systems, and trade networks can be restricted in ways that compel behavioural change. This premise was historically valid within a more centralised global system, where key economic pathways were concentrated and dependencies were clearly defined. In the contemporary environment, however, this assumption is increasingly misaligned with reality, as the global economy has evolved into a distributed network capable of adapting to constraints rather than being halted by them.
The effectiveness of sanctions depends fundamentally on the existence of enforceable control points, which historically included dominant currencies, centralised banking systems, and established trade routes that could be monitored and regulated with relative precision. The influence of the United States over the global financial system, particularly through the dollar, has long provided a mechanism for imposing such controls, enabling policymakers to restrict access to capital and thereby exert pressure on targeted states. This framework assumes that participation in these systems is essential and that exclusion carries sufficient cost to influence decision-making.
As the global economy has diversified, the exclusivity of these control points has diminished, not through deliberate dismantling but through the gradual emergence of alternative pathways that allow actors to operate outside traditional frameworks. Iran’s response to sustained sanctions illustrates this evolution, as it has developed a sophisticated network of logistical and financial mechanisms designed to maintain the flow of oil despite formal restrictions. These mechanisms include ship-to-ship transfers conducted in international waters, the use of intermediary entities to obscure ownership and origin, and the reclassification of cargo to facilitate entry into markets that would otherwise be inaccessible.
These adaptations are not isolated or temporary measures but have become institutionalised practices that form the basis of a parallel economic system, one that mirrors the official market while operating under different rules and levels of transparency. The existence of this system demonstrates that sanctions do not eliminate economic activity but instead alter its structure, redirecting flows into channels that are more complex and less visible. This redirection introduces inefficiencies, including higher transaction costs and increased logistical complexity, yet these costs are absorbed within the system as long as underlying demand remains strong.
Russia’s experience further reinforces this pattern, as sanctions imposed in response to geopolitical conflict have not resulted in a cessation of exports but rather in a reorientation toward markets willing to engage under alternative conditions. By redirecting oil flows to countries such as China and India, Russia has maintained its revenue streams while simultaneously reducing its reliance on Western markets. This reconfiguration is facilitated by the same types of mechanisms observed in Iran’s case, including adjusted payment systems, intermediary trading arrangements, and logistical adaptations that enable continued participation in global trade.
The persistence of these flows highlights the limitations of enforcement, which is inherently constrained by the complexity and scale of the global system. Monitoring and regulating every transaction, shipment, and financial exchange is not feasible, particularly when actors are incentivised to develop increasingly sophisticated methods of circumvention. Each new layer of restriction generates a corresponding layer of adaptation, creating a dynamic in which enforcement and evasion evolve in parallel without reaching a definitive resolution.
Currency diversification represents a critical component of this adaptation, as transactions conducted outside the dollar system reduce the effectiveness of sanctions that rely on financial exclusion as a primary tool. Bilateral agreements, the use of alternative currencies, and the development of new payment mechanisms all contribute to a gradual reduction in dependency on centralised financial infrastructure. While these alternatives do not yet match the scale and efficiency of established systems, their continued expansion reflects a broader trend toward fragmentation, where multiple pathways coexist rather than a single dominant framework.
The redistribution of cost within sanctioned environments is another factor that complicates the intended impact of these measures, as economic pressure is often diffused across populations rather than concentrated on decision-makers. Consumers experience higher prices, businesses encounter disruptions, and workers face reduced opportunities, while political leadership may be insulated from these effects through internal mechanisms or alternative revenue streams. This diffusion of impact can weaken the direct link between economic pressure and political change, reducing the likelihood that sanctions will achieve their stated objectives.
In some cases, external pressure may even reinforce internal cohesion, as sanctions are framed as unjust or politically motivated, thereby strengthening narratives of resistance and sovereignty. This dynamic illustrates the importance of perception in determining the effectiveness of policy tools, as the interpretation of sanctions can influence whether they produce compliance or defiance. The variability of outcomes across different contexts underscores the complexity of applying uniform measures to diverse political and economic systems.
The cumulative effect of these dynamics is the gradual transformation of the global economy into a more fragmented and less transparent network, where parallel systems operate alongside official ones and where control is distributed rather than centralised. This transformation does not occur through a single event but through a series of incremental adjustments that collectively reshape the architecture of trade and finance. Each adaptation, while individually limited in scope, contributes to a broader shift that reduces the effectiveness of traditional policy tools.
Sanctions, within this context, function less as instruments of control and more as catalysts for innovation, prompting the development of new systems that operate outside established frameworks. This unintended consequence highlights the importance of aligning policy tools with the structure of the system in which they are applied, as measures designed for a centralised environment may produce different outcomes in a distributed one.
The illusion of sanctions lies not in their complete ineffectiveness but in the discrepancy between their intended purpose and their actual impact, as they often produce structural changes that extend beyond their immediate objectives. Understanding this discrepancy is essential for developing more effective approaches to economic policy, as it requires recognising the adaptive capacity of global systems and the limitations of control-based mechanisms within a fragmented environment.
For policymakers, this means re-evaluating the role of sanctions within a broader strategy that incorporates alternative tools capable of influencing behaviour in more nuanced ways. For businesses, it highlights the importance of navigating a landscape where regulatory environments are increasingly complex and where supply chains must be resilient to both formal restrictions and informal adaptations. For the global economy, it underscores the transition toward a system characterised by multiple overlapping networks, each with its own rules and dynamics.
The persistence of trade under sanction conditions demonstrates that economic systems are not easily halted, but rather reconfigured, and that efforts to impose control must account for this capacity for transformation. Recognising the limitations of sanctions does not imply abandoning them, but it does require a more sophisticated understanding of how they interact with the systems they are intended to influence.

Artificial intelligence arrives on our screens almost without weight. A sentence materialises. An image appears. A model reasons through a problem in seconds. The interface encourages a seductive fiction: intelligence has escaped matter. It lives somewhere called the cloud. The economics now reveal the opposite. AI is becoming one of the most physically demanding capital projects of the modern era. In April, the International Energy Agency reported that capital expenditure among five large technology companies exceeded $400 billion in 2025 and was expected to increase by another 75 per cent in 2026. This month, Nvidia announced arrangements with major financial institutions intended to mobilise more than $500 billion of third-party capital for AI infrastructure. Alphabet, meanwhile, has returned repeatedly to debt markets as technology companies finance an AI investment cycle that Reuters says could push sector spending beyond $730 billion this year. Money is only the beginning. Intelligence at industrial scale requires semiconductors, servers, transformers, substations, transmission networks, cooling equipment, water, land, concrete, skilled labour and — above everything — electricity. Data-centre electricity demand rose 17 per cent in 2025, according to the IEA, while AI-focused facilities grew faster still. The agency now expects data-centre electricity consumption to double by 2030, with electricity use at AI-focused centres potentially tripling. Then comes the environmental contradiction. A Financial Times analysis of 60 large planned American data-centre projects estimates potential annual emissions of approximately 101.5 million tonnes of carbon dioxide if their projected electricity requirements are supplied under anticipated generation conditions. Utilities are adding gas capacity, and some coal retirements are being delayed as electricity demand accelerates. We called it artificial intelligence. The infrastructure required to produce it is brutally physical. The consequential AI story is therefore no longer merely which model can reason fastest, generate the best video or dominate the next benchmark. The deeper story is the emergence of an industrial system capable of reorganising capital, electricity, land, supply chains and geopolitical power around the production of machine intelligence. The cloud has touched the ground. And what it is building there may prove considerably more important than the chatbot.

The internet has already written the spectacular version of this story: Japanese scientists have reversed ageing, discovered an anti-ageing drug, and opened the possibility that humans could live for 250 years. The actual science is both narrower and more interesting. Researchers at the University of Osaka identified a protein called AP2A1 that appears to help maintain some of the enlarged structural characteristics of senescent cells. When researchers suppressed AP2A1 expression in ageing human fibroblasts, several characteristics associated with cellular senescence were reversed: cells became smaller, recognised senescence markers decreased, and proliferative and migratory activity increased. When AP2A1 was overexpressed in younger cells, senescence-associated characteristics advanced. The peer-reviewed study was published in Cellular Signalling in January 2025. That is a legitimate scientific finding. It is not evidence that human ageing has been reversed, nor that humans could live for 250 years. The study was conducted principally in cultured human fibroblast and epithelial cell models. It did not demonstrate age reversal in a human being, establish lifespan extension, or test a treatment capable of producing radical human longevity. Yet dismissing the research because social media exaggerated it would make the opposite mistake. The consequential finding is that cellular senescence may be more mechanically configurable than previously understood. Ageing cells do not simply accumulate molecular damage; their physical architecture may actively help maintain the senescent state. That changes the question. Perhaps ageing is not merely something cells endure. At least some characteristics of cellular ageing may be states that biological systems actively maintain — and therefore states that science may eventually learn to modify.

Two sisters living with an exceptionally rare growth disorder have become part of a much larger scientific question: what if one of the most useful ways to understand cancer is to study human bodies in which some of the biological conditions cancer exploits have been altered from birth? The condition is Laron syndrome, a form of growth-hormone insensitivity caused by dysfunction of the growth-hormone receptor. The resulting biology includes extremely low levels of insulin-like growth factor 1, or IGF-1, an important regulator of growth, metabolism, and cellular activity. Researchers studying people with Laron syndrome, particularly a distinctive cohort in Ecuador, have repeatedly reported unusually low incidences of cancer and diabetes compared with relatives and wider populations. The finding has attracted scientific attention for decades. But “The Cancer Twins” must resist the temptation that makes the story irresistible online. Laron syndrome does not establish human immunity from cancer. Cancer has occurred within the Ecuadorian population, and the evidence concerns substantially diminished incidence, not biological invulnerability. A 2023 review of the Ecuadorian cohort explicitly documented cancer cases while maintaining the larger finding of unusually low cancer incidence. The more consequential proposition lies underneath the headline. Cancer requires more than mutations. Malignant cells must obtain signals, energy, metabolic support, opportunities to proliferate, and mechanisms for avoiding destruction. By studying people whose growth-signalling architecture differs markedly from the norm, researchers may gain an unusual natural experiment into the conditions that make malignant growth easier — or harder. The twins, then, are not scientific curiosities. Nor are they miracle patients. They represent something more intellectually valuable: a human biological exception capable of revealing the rules governing the rest of us.