The Rulebook Is Being Rewritten, and Not Only in Washington.

 

The Rulebook Is Being Rewritten, and Not Only in Washington.

The End of the Repair Era.

Something fundamental has shifted in how the world's most powerful nations approach their own institutions. For decades, the operating assumption of Western statecraft was that the post-war order could be fixed, patched, and improved through incremental reform. That assumption is now collapsing under the weight of its own failures. A growing share of voters and political movements across Western democracies have stopped believing the system can repair itself. Instead of demanding fixes, they are choosing demolition. Institutions are being dismantled not because replacements are ready, but because faith in the old machinery has evaporated. This is not a policy shift. It is a deeper structural realignment—a crisis of legitimacy that no election cycle can resolve.

The distinction matters because it explains why conventional analysis keeps missing what is happening. Policy shifts respond to incentives. Legitimacy crises respond to narrative. When citizens conclude that the institutions meant to serve them are beyond correction, they do not vote for better managers. They vote for wreckers. The emotional logic is not irrational. It is a rational response to decades of promises that never materialized. Wages stagnated while elites prospered. Wars were launched and never won. Crises were managed in ways that protected the connected and punished the exposed. The cumulative effect is a public that no longer trusts the machinery to deliver, and therefore no longer cares whether it survives.

The danger in this approach is that demolition is not a plan. It is a release valve. When principles give way to transactions, when shared interests are subordinated to private ones, and when regional power replaces universal rules, the result is not liberation. It is drift. The world that emerges from this period will not be shaped by those who tear down fastest, but by those who build while others are distracted. History is unkind to nations that mistake destruction for renewal. The powers that endure are rarely the loudest in the room. They are the ones quietly laying foundations while everyone else argues about the furniture.

There is a further irony worth naming. The very nations now questioning their own institutional order are the same ones that spent decades exporting that order to others as a universal model. When the model's architects lose faith in it, the export market collapses. Rising powers no longer need to argue that the Western system is flawed. They can simply point to the West's own behavior and let the evidence speak. The result is not a rival ideology triumphant, but a vacuum of authority. And vacuums do not stay empty for long. They are filled by whoever is willing to set terms rather than ask permission. That is the real story of this moment. The rulebook is not being rewritten in Washington. It is being rewritten everywhere else, while Washington argues about whether the rulebook should exist at all.Technology as a Border Wall.

Technology as a Border Wall.

The clearest expression of this new era is technology policy. Restrictions on advanced robotics and research institutions are being justified in the language of national security, but the pattern tells a different story. When limits are aimed so precisely at one country's consumer products and university labs, the security rationale starts to look like a cover for industrial protection. The question is not whether great powers should protect their technological advantages. The question is whether they can do so without mistaking a fence for a foundation.

What makes this moment distinct is not the existence of technology restrictions. Great powers have always guarded their advantages. What is new is the collapse of the distinction between security and commerce. When every trade measure becomes a security measure, and every security measure becomes an industrial policy, the categories lose meaning. Allies cannot tell whether they are partners or targets. Rivals cannot tell whether they are negotiating or preparing for conflict. The result is a system where no one knows the rules because the rules change with the political weather. That uncertainty carries its own cost. It discourages the long-term investment that technological leadership actually requires.

History offers a warning here. Great powers rarely decline because a rival grows stronger. They decline when their own politics can no longer reform. Pointing at an external threat can delay that reckoning, but it cannot replace the hard work of self-correction. A fence line may keep out a competitor. It will not keep out decay. The Soviet Union did not collapse because the United States out innovated it. It collapsed because its own system could not adapt. Rome did not fall because barbarians were stronger. It fell because its institutions had calcified beyond repair. The pattern repeats because the lesson is uncomfortable. External threats are easier to name than internal failures. Enemies are more satisfying to blame than decay.

There is a deeper problem with using security as a blanket justification. It removes the need for evidence. Once a policy is labeled a security matter, it becomes immune to cost-benefit analysis, immune to diplomatic negotiation, and immune to democratic oversight. The classification itself becomes the argument. This is not how healthy institutions function. It is how declining ones avoid accountability. A nation confident in its own trajectory does not need to label every competitor a threat. It competes. It innovates. It wins. A nation that has lost confidence in its own model reaches for restrictions because it has run out of better ideas.

The AI Race Has Changed Its Currency.

The artificial intelligence contest has entered a new phase. Raw capability used to be the headline. Now cost-effectiveness decides who wins. On quality, the gap between the best American and Chinese models has narrowed to the point of being nearly meaningless. When a lead lasts only a season, it stops functioning as a moat. What matters now is price, and the price difference is not incremental. It is staggering.

This shift has profound implications that most analysis misses. When quality converges, the competitive advantage moves to whoever can deliver acceptable performance at the lowest cost. That is not a technology story. It is an infrastructure story. It is about electricity prices, data center efficiency, chip supply chains, and the ability to scale without bottlenecks. The nation that wins the AI race may not be the one with the best models. It may be the one with the cheapest power and the fastest permitting process.

Buyers are responding accordingly. Open-source models are capturing a rapidly growing share of processed tokens. Institutions are migrating from proprietary systems to open alternatives in large numbers. Chinese open models have become a common base layer for developers around the world, and a striking share of American AI startups now mention them in their pitches. If that is true, then "decoupling" sounds less like a description of reality and more like a slogan. The products people build tell a different story than the speeches people give.

There is an irony here that deserves naming. The push for technological separation happens precisely when technological integration is accelerating. Developers do not care about geopolitical boundaries when they are choosing a model. They care about cost, performance, and reliability. If an open model from one country performs as well as a proprietary model from another at a fraction of the price, the developer will choose the cheaper option regardless of where it was built. This is not a failure of patriotism. It is a feature of rational decision-making. And it means that attempts to wall off technology from competition will fail not because they are immoral, but because they are impractical.

The deeper lesson is that technology does not respect the boundaries that politics draws. It flows around obstacles. It finds the path of least resistance. A nation that tries to monopolize innovation will find itself isolated not from rivals, but from the collaborators it needs to stay competitive. The future belongs not to those who build the highest walls, but to those who build the widest bridges.

The Infrastructure Question.

Power Moves Below the Leaderboard.

Leaderboards will not settle this contest. Chips, electricity, data centres, and cloud regions will. The ability to finance, power, and permit infrastructure may matter more than the quality of the models that run on it. This is a lesson that the technology industry learns slowly and forgets quickly. Software scales at the speed of imagination. Infrastructure scales at the speed of politics, capital, and land. The first is abundant. The second is not. That asymmetry is what separates a demo from a dominion.

There is a deeper point buried in this shift. For two decades, the assumption was that innovation flows from the top down, from elite research labs in a handful of countries to everyone else. That assumption is being inverted. The countries that can move electricity, land, and permits fastest are often not the ones with the most prestigious research institutions. They are the ones with the most functional states. Infrastructure is not a technical problem. It is a governance problem. And governance is now a competitive advantage in its own right.

Africa as the New Centre of Gravity.

This puts regions that have long been treated as peripheral at the centre of the story. Africa offers a case in point. Adoption numbers are remarkable by any measure. In Kenya, nearly every adult internet user engages with AI tools monthly. South Africa and Nigeria are not far behind. In Nigeria, a large majority use AI weekly and consider themselves capable users, far above the global average. These are not passive consumers. They are active participants in a technological shift that is happening faster than most analysts predicted.

What makes these numbers significant is not their size but their character. Adoption this rapid, in markets with limited formal infrastructure, suggests something that development economists have long argued about but rarely proven: leapfrogging is real. Mobile money proved it in finance. AI adoption is proving it again in productivity. The absence of legacy systems is not a handicap. It is a clean slate. Countries that never built extensive branch banking moved directly to phones. Countries that have never built extensive desktop computing are moving directly to AI-assisted work. The implication is uncomfortable for incumbents. The next wave of AI users will not look like the last wave of PC users. They will be mobile-first, price-sensitive, and unwilling to pay for prestige.

Two Offers, One Choice.

Two competing offers are on the table. One packages cloud and AI services with affordable devices and digital payment systems, arriving as one more layer on top of what people already use. The other asks applicants for full-stack bids covering hardware, cloud, networks, and applications, with a significant share of value required to come from domestic sources. The choice between them is not just about technology. It is about dependency, leverage, and who gets to set the terms.

The first offer is built on integration. It assumes that adoption matters more than control, and that being embedded in people's daily habits is a form of influence that outlasts any contract. The second is built on conditionality. It assumes that dependency is a risk to be managed, and that local value creation is worth a higher price. Both have merit. Both have costs. The first risks lock-in to a foreign stack. The second risks slower deployment and higher prices. The real question is not which offer is better in the abstract. It is which offers a given state that has the capacity to negotiate. That is where most of the Global South's leverage quietly disappears. Not because the offers are unequal, but because the negotiating capacity is.

This asymmetry deserves closer examination because it is the hidden mechanism that determines outcomes. Integration-based offers are seductive precisely because they require so little from the recipient. They arrive ready-made, priced for mass adoption, and designed to work immediately. The cost is deferred, not eliminated. Lock-in happens slowly, through habit and interoperability, until switching becomes unthinkable. Conditionality-based offers demand more upfront—legal review, technical assessment, political will—but they preserve optionality. The state that can absorb that complexity gains something the other state never will: the ability to change its mind.

There is a third path that rarely gets named. A state can accept integration now while building the capacity to renegotiate later. This is not a compromise between the two offers. It is a sequencing strategy. Use the accessible offer to accelerate adoption, then use the resulting dependency as leverage when the terms come up for renewal. The risk is that renewal never comes, or that the dependency deepens faster than the capacity to renegotiate. The reward is that a country does not have to choose between being connected and being sovereign. It can be both, if it plays the sequence correctly.

The deeper problem is that most states in the Global South are not playing a sequence. They are playing catch-up, one deal at a time, with institutions that were not built for this kind of negotiation. Trade ministries are understaffed. Energy regulators are underfunded. Data protection authorities are underpowered. The offers are not designed to exploit this weakness. They simply flow into it, the way water flows into the lowest ground. Fixing requires investment in people, not just infrastructure. A country can have the fastest data centre on the continent and still be a rule-taker if no one in the room knows how to price a concession or read a contract.

This is why the negotiating capacity gap is more consequential than the offer gap. Offers can be improved. Terms can be revised. But a state that lacks the ability to negotiate will accept whatever is presented, and it will accept it again the next time. The cycle repeats until someone breaks it. Breaking does not require rejecting foreign investment. It requires being prepared enough to shape it. That is the real choice on the table. Not between two offers, but between two futures: one in which the terms are set elsewhere, and one in which they are set at home.

What a Stalled Project Reveals.

A delayed data-centre project in Kenya offers a useful lesson. The project stalled not because technology or money was unavailable, but because the government was unwilling to give the long-term financial guarantees and energy commitments the deal required. What limited progress was not a lack of offers. It was the state's ability to absorb the deal on its own terms.

The implication is that governments in the Global South hold more cards than the old chessboard metaphor suggests. Countries with capable regulators and skilled negotiators can turn competition into leverage. Staying neutral between competing blocks and drawing on both is a viable strategy. The key is tying infrastructure deals with industrial policy and local skills, so that the terms of engagement are set domestically rather than externally.

There is a hard truth here that deserves emphasis. The difference between a country that benefits from great-power competition and one that is exploited by it is not geography or resources. It is institutional depth. A state that cannot model a deal, cannot price a guarantee, and cannot enforce a contract will always be the junior partner, regardless of how many suitors come calling. This is why capacity building is not a soft issue. It is the hardest infrastructure of all. Data centres can be built in two years. A competent negotiating team takes a generation.

From Being Represented to Representing Themselves.

A slower but equally significant shift is under way. Global South is moving from a label to an actor. Expanded alliances now cover a substantial share of global GDP and nearly half the world's population. Intra-bloc commodity exports have grown dramatically, and development banks have approved tens of billions of dollars across hundreds of projects. These are not symbolic numbers. They represent a real reorientation of economic gravity.

But numbers alone do not make a block. What makes a block are shared rules, and shared rules require shared enforcement. This is where the story becomes more complicated than the headlines suggest. Many of these alliances are internally diverse, with members that compete as much as they cooperate. The glue is not ideology. It is a common frustration with being excluded from decisions that affect them. That frustration is real, but it is also a thin foundation. Frustration can bring countries to the table. It cannot keep them there. Only tangible benefits can.

The Ambition to Write Rules.

At recent summits, proposals have focused on open and inclusive AI, trade and investment facilitation, digital industry cooperation, smart manufacturing, and science and technology talent. The ambition behind them is to move from having a voice to writing rules. This matters because of how AI governance may develop. When two major capitals agree to a bilateral dialogue, the risk rises that two sets of interests will shape the rules for everyone. Countries left out could end up with standards built for other people's needs. An alternative track based on openness and low cost offers a way to avoid being a permanent rule-taker.

The strategic logic here is sharper than it first appears. Rule-taking is not just a matter of pride. It is a matter of cost. Standards determine which products can be sold, which data can be moved, and which companies can compete. A country that imports its standards imports its economic structure along with them. Open-source approaches matter precisely because they lower the barrier to participation. A country that cannot afford a proprietary AI stack can still build on an open one. That is not charity. It is a different model of influence; one based on adoption rather than control. Whether it works will depend on whether openness is matched by reliability, security, and sustained investment. Openness is a starting point. It is not a strategy on its own.

A Multipolar World Is Not a Stable One.

The world is now multipolar. That much is widely accepted. But multipolarity describes the breakdown of a Western-led system, not the arrival of a replacement. The fault lines are plain. A war of attrition in Europe whose direction may be decided as much in Washington as on the battlefield. A fragile ceasefire in the Middle East with elections as a possible disruptor. Border disputes in South and Southeast Asia sharpened by religious and territorial tensions. And an economy where global growth continues but trade grows only marginally, a sign that integration is giving way to fragmentation.

Multipolarity is often described as if it were a destination. It is not. It is a transitional condition, and transitional conditions are the most dangerous kind. Bipolarity had rules, however harsh. Unipolarity had a referee, however partial. Multipolarity has neither. It has multiple centers of power, overlapping interests, and no agreed mechanism for resolving disputes. That is not stability. It is a system held together by mutual caution and temporary convenience.

The Costs of Fragmentation.

Fragmentation carries costs, and everyone pays some of them. The question is not whether the old powers will wake from their slumber. It is what they do when they do. If they continue demolishing, the damage will not stay on the other side of the wall. For Africa, China, and India, the work is not to wait and respond. It is to build now, so that when the old powers open their eyes, they find a world they can no longer define alone.

The deeper insight is that fragmentation is not evenly distributed. It is felt first and hardest by those with the least cushion. Small economies that depend on trade rules they did not write suffer most when those rules erode. Large economies can absorb shocks, retaliate, and set their own terms. The gap between the two widens in periods of disorder. This is why Global South's push to write rules is not an act of ambition. It is an act of self-preservation. The alternative is to be perpetually rule-taken in a world that no longer agrees on what the rules should be. Building now is not a slogan. It is the only rational response to a system that is rewriting itself whether anyone is ready or not.