What phase is the global system in now?

How expansion gave way to limits, and why that now feels normal

Large systems rarely move in straight lines, but they do shuffle along, expanding, experimenting, sometimes overreaching, absorbing shocks, and eventually consolidating into constituent parts. The ever-expansionist, high-velocity phase that characterized much of the post-Cold War period, which was largely defined by rapid integration, permissive governance, and an unspoken faith in self-correcting markets, was always going to give way to something more constrained.

At present, it feels like we’ve reached that kind of inflection point, and what we are living through now is not a sudden break, but the consolidation phase, the point at which accumulated risks, dependencies, and institutional lag force systems to narrow their operating range.

If we choose to be reactive, to be deluged by the neverending stream of news, upheaval of ‘norms’, the sensationalization of the world, then this phase is often misread as outright crisis, or some kind of reversal of progress. In practice, however, it is something closer to normalization, for expansion creates optionality, while consolidation reduces it, and the political and economic work of consolidation is not about inventing new rules, but about deciding which provisional arrangements become permanent, the ongoing manner in which we choose to cooperate.

This is perhaps why this moment feels quieter than one of outright disruption, for the debates are procedural, technical, and often framed as “maintenance” rather than “transformation”. Their consequences, however, are durable.

Why expansion could not last

Since the mid-1990s, globalization and technological growth have rested on a set of assumptions that now, in retrospect, look a little fragile. Efficiency was treated as an unquestioned good, the quest for it evident in the unbridled industrial relocation to a “cost-effective” China. Governance was seen as something that could safely lag innovation, for as long as the economic wheels spun ever faster, the rules would follow. And, within this activity we saw interdependence as a force that would reduce coercion, blunting it, rather than actually shifting it. Ok, sure, for a time, those assumptions paid off: trade expanded, capital moved with few barriers, supply chains optimized, and digital systems scaled across borders with a limited degree of oversight. The gaps in governance were visible but politically tolerable, because the potential failures still seemed manageable and, above all, reversible.

Over time, however, that implicit trust in reversibility eroded. Financial crises demonstrated how tightly knitted systems could transmit shocks faster than institutions could respond, while energy markets revealed that efficiency-driven integration just created new vulnerabilities once the underpinning geopolitical trust started to weaken. Digital platforms expanded into labor markets, information systems, and public administration at a scale that overwhelmed the negligible levels of real oversight. Each episode has produced a suite of targeted fixes, the stress tests, mandatory reporting requirements, or just voluntary codes, but these accumulated faster than the institutional capacity necessary to absorb them. Then, by the early 2020s, expansion no longer added resilience, rather, it increased exposure.

What ultimately broke the logic of expansion was not any single failure, but the convergence of risks across domains that had previously been managed separately. Computing power, or compute, scaled alongside capital concentration, turning infrastructure into leverage rather than a neutral input. Energy demand tied digital growth directly to grid stability and climate politics. Model capability advanced faster than interoperability, pushing liability and accountability upstream. In that environment, “fixing it later” ceased to be credible not as a moral claim, but as an operational one. The costs of reversal began to exceed the costs of pre-emption, forcing systems to narrow their own range of acceptable behavior.

And so, we start to feel a slight squeeze.

We’ve seen consolidation before

This pattern isn’t new. Periods of rapid expansion are often followed by phases where the rules have to catch up. For example, global finance liberalized quickly in the late twentieth century, then was wrapped in layers of regulation after the shock of the 2008 crisis. Likewise, post-war energy markets expanded, only to be followed in the 1970s by strategic reserves, a morass of cartel politics, and doctrines based around the security of access and supply to oil. Even the early internet, long defined by openness and minimal oversight, has of late accumulated more standards, liability rules, and national controls.

In each case, consolidation did not undo what came before; instead, it absorbed it. Systems that grow faster than their ability to be governed eventually end up narrowing what is permitted. Flexibility gives way to predictability, a shift that is rarely presented as ideological. It is justified as “risk management “and “stability”, and once those rules are in place, reversing them becomes both politically costly and increasingly unlikely.

Read: 2025: when technology stopped being a free-for-all

Where we now see consolidation

This consolidation phase is already visible across several areas. In technology, rules that were once tentative have become part of the baseline operating approach. For example, the European Union’s AI Act does not just respond to specific abuses, but sets boundaries in advance by defining which uses are allowed and which are not, and by forcing those limits into system design from the start. The significance of this is pretty straightforward: some technologies are now being shaped before they are deployed, rather than corrected afterward (EU AI Act).

Trade and industrial policy show the same shift. “Economic security” has become a common organizing idea, where supply chains, technology, and geopolitics are tied together in ways that would perhaps have been controversial a decade ago. Export controls, investment screening, and subsidies are no longer treated as emergency measures, but as the standard tools of government policy. Energy policy, similarly, has followed this path, moving away from pure efficiency toward the necessity for hedging and ensuring backup capacity, with redundancy redefined as “protection” rather than simply as waste (European Commission).

It should be noted that none of this points to collapse or isolation. No headless chickens, scrambling to make sense of the chaos. Instead, trade continues, capital is still allocated where it needs to go, and innovation certainly has been stopped in its tracks. However, what has changed is the range within which systems are expected to operate, for they are now built on the assumption that limits do indeed exist, and this is an assumption that then shapes decisions all the way downstream.

The narrowing of acceptable choices

A key feature of consolidation is that choices narrow without anyone formally announcing it, which makes sense if you consider that “limiting your options” never made for a healthy marketing tagline. Measures introduced as temporary responses in the face of overt risk tend to quietly stick around once people have started to comfortably adapt to them: companies reorganize their operations around new rules; regulators build teams to enforce them; legal interpretations of what is happening start to pile up; and partners abroad adjust how they deal with each other. Over time, what was meant to be a short-term provisional fix becomes part of the furniture, the system, largely because removing it would require a deliberate show of political effort rather than simply letting it lapse.

This helps explain why the current phase feels settled even without experiencing a single, impactful, defining event. Consolidation inches forward through routine use, not via loud declarations. If you think about it, export controls, often introduced as safeguards, become standard screening tools, the interim becoming basic and accepted. Another example is the perceived pain of implementing sustainability reporting rules, yet they quickly and quietly tend to reshape how organizations manage themselves, how they see how they can manage themselves. Each step looks small on its own, but together they make the operating environment more rigid, and by the time the effects are obvious, the logic that constructed them is already built into budgets and institutions, the way we do business as usual.

This also changes where political conflict shows up, since the argument is no longer about whether limits are needed, but about which limits are acceptable and who pays the price for them. It’s not that disagreement has totally disappeared, rather, it’s more the case that this has shifted into technical processes and squabbles over procedure. So, once systems are built on the assumption that limits do indeed exist, that ever-expansive sense of openness stops being the unspoken default, instead becoming conditional, selectively applied, and, as circumstances change, withdrawn.

Why does this phase feel uncomfortable?

Consolidation is often experienced as loss: of flexibility, of speed, of an imagined possibility of futures. Conversely, during expansion, possibility feels like our natural everyday companion, both wholly abundant and most definitely reversible, should it be warranted. During consolidation the tradeoffs are no longer hidden and systems focus more on simply keeping things running than on chasing the next new idea. It’s some form of a ‘flight’ mode, dealing with apparent crisis or the intimation of one. Political decisions reflect this tendency, starting to interfere with and ultimately shape areas that were once the ungoverned province of the ever-self-correcting markets. That is an uncomfortable tension, sure, but it does reflect systems adjusting to their own size and limits of operation, it’s not a breaking down of functioning.

The risk here is not recognizing consolidation, but misunderstanding it. Treating it simply as a temporary anomaly, an aberrant blip on the predictable procession of things, this encourages an overreaction or sense of nostalgia for a past that isn’t coming back. If we instead treat it as the end of the road, the ultimate narrowing to the point of constriction, then this leads to passivity, to our inactivity or outright apathy. In reality, consolidation phases are periods when the rules for the next phase are being formulated and set, taking their shape often unevenly, and in ways that tend to favor the established players.

What attention should focus on now

If expansion was all about ebullient growth and consolidation is about the move toward stability, the question now is not how to reopen everything, but how exactly the limits are being set. Which risks are taken seriously? Who will pay for them? Who gains real influence as systems harden into their new shapes? This is what comes next.

This phase favors attention to infrastructure, to the rules and institutions rather than to the blaring, sensationalized headlines. Politics hasn’t disappeared from technology, either, it has just moved over into the standards, liability, and control over key systems. You don’t need to be the eternal pessimist to understand that, you just need to develop a sense of how these cycles usually play out.

So, perhaps seeing the present world, where we are right now as a consolidation phase, perhaps this can help us avoid both veering toward outright panic, or settling into complacency? It can explain why so many developments do feel familiar, even when their details change, and why the most important decisions often arrive, somewhat clandestinely, dressed up as technical adjustments that ultimately carry larger weight.

Nope. Not the end of openness. Also nope: not the start of something entirely new. We’re in the middle of a cycle, and what matters now is which constraints are already locked in and which are still to be contested?

Read this.

The OECD’s AI Policy Observatory on upstream governance and accountability; the International Energy Agency’s analysis of data-center energy demand and grid constraints; and Stanford’s AI Index Report on compute concentration and capital intensity.

Notice that.

None of these sources argues that systems are breaking! Instead, they document how accumulated risk forces rules to catch up, and this is often in incremental, procedural ways that may only look “inevitable” in hindsight.

Do something.

Shift attention from novelty to structure. Watch which provisional measures stop being debated and start being assumed. That is where consolidation becomes durable and, importantly, where leverage still exists.

Previously on GYST: Water as constraint: scarcity, leverage, and the geopolitics of survival