The Illusion Of Momentum.

In an era defined less by risk than by ambiguity, the most dangerous momentum may be the one we believe we have.

Across the world, geopolitical fractures widen without warning. Once-peripheral technologies (like generative AI) reconfigure entire industries in quarters, not decades. Climate constraints redraw cost curves and capital flows. While the surface narrative is turbulence, the deeper pattern is something more subtle and dangerous: the illusion of momentum.

Organizations move. Dashboards update. Strategies roll forward. Revenue grows, until it doen’t. From the inside, this motion feels like progress. It feels directional. It feels earned. Businesses interpret this forward motion as evidence of underlying stability, as if the system itself is solidifying their trajectory. They call it traction, resilience, execution…

However,  what if much of that „momentum” is simply inertia encountering temporary tailwinds?

The illusion of momentum occurs when businesses mistake continuity for causality. When recent performance is read as proof of sustainable advantage. When volatility is framed as noise rather than signal. At such times, uncertainty is domesticated, reclassified as risk. And risk, unlike uncertainty, feels manageable. It can be modeled, priced, hedged, optimized. It suggests that with enough data, enough analysis, enough refinement, clarity will emerge.

This is the misdiagnosis.

Risk assumes known distributions. Uncertainty does not. Risk implies calculable odds. Uncertainty implies shifting realities. Yet in boardrooms and strategy offsites, the language of probability quietly replaces the language of ambiguity. Leaders reach for sharper  forecasts, denser dashboards, tighter KPIs,  believing that resolution lies just beyond the next analytical iteration.

Cognitive traps reinforce the pattern. Narrative fallacy tempts decision-makers to adapt coherence to randomness. An explanation appears plausible, elegant, reassuring, and the mind begins its confirmatory search. Data is filtered. Dissent is softened. Alternative futures fade from view. What feels like disciplined reasoning is often disciplined storytelling.

Under the illusion of momentum, organizations double down precisely when they should diversify. Optimize exactly when they should explore. Scale precisely when they should be questioning the very premise of scaling.

The first error is not analytical, but conceptual. It lirs not in the spreadsheet, but in the frame. When deep uncertainty is mislabeled as manageable risk, the tools deployed (forecasting models, scenario ranges, incremental investments) become misaligned with reality. Predictable mistakes follow: overcommitment to a single path, underinvestment in optionality, delayed recognition of regime shifts.

That’s why many business failures today are not failures of intelligence or effort, but failures of diagnosis. Leaders are not blind to volatility, they are blind to the type of volatility they face. They misinterpret uncertainty as momentum and momentum as evidence.

Let’s explore further how organizations in the AI ​​era can navigate the illusion of momentum, operate within predictable unpredictability, and design strategies that transform ambiguity from threat to opportunity.

In 1927, German physicist Werner Heisenberg formalized what became known as the uncertainty principle, one of the foundations of quantum mechanics. It states that certain pairs of properties, such as position and momentum, cannot both be known with perfect precision at the same time. The more precisely one is measured, the less precisely the other can be determined.

This is not a limitation of instruments, but rather a property of small-scale reality. The act of measuring one variable affects the system in a way that constrains our knowledge of the other. For example, we know that thunderstorms will form under certain atmospheric conditions, but predicting the exact timing, intensity, and lightning strikes is much more difficult.

We know that stock markets, such as the S&P 500 trend upward over long horizons, yet daily movements remain volatile and difficult to forecast. We know that economic cycles repeat, but recessions rarely occur when the consensus expects them. We know that real estate markets move in waves, but bubbles and crashes appear obvious only in hindsight. And so on…

Today, AI has entered the mainstream of boardroom conversations at breakneck speed. Investment flows are accelerating. Capabilities are expanding. Competitive pressure is intensifying across industries once considered mature and stable. The prevailing narrative suggests that the momentum is obvious and direction is clear.

Yet this perception conceals a deeper reality: AI’s future will look inevitable in 2035. In 2026, it is still ambiguous… And this ambiguity is not a flaw in analysis. It is not a temporary fog waiting to lift. It is the terrain itself that we must navigate:

• Momentum is visible, yet not linear.
• Direction is probable, yet not precise.
• Dominance is plausible, yet not guaranteed.

Businesses that understand this distinction between structural inevitability and tactical uncertainty will navigate ambiguity with confidence, rather than fear. And those who learn to operate within ambiguity, rather than wait for it to disappear, will be the ones who shape the outcome that history later deems inevitable.

Here are some perspectives to consider:

I. The return of creative destruction

The unsettling business headlines of today are not isolated shocks. They represent the ongoing expansion of Joseph Schumpeter’s concept of creative destruction into sectors long considered mature and defensible.

For decades, many businesses  operated under conditions of relative stability. Barriers to entry were high. Transaction costs limited disruption. Information flowed slowly. Competitors were known and finite. Markets expanded steadily in increasingly affluent economies. Strategy could be optimized around sustainable advantages and forecastable growth.

Leadership mindsets evolved accordingly. Financial models, such as net present value, assumed that future cash flows can be estimated with reasonable accuracy and discounted back to present value. Planning cycles have emphasized optimization over exploration. Competitive gaps have been treated as structural and defensible.

Nowadays, that operational environment has changed (or is about to change) fundamentally.

Meanwhile, digitalization reduced transaction costs. Cloud infrastructure lowered capital intensity. Global platforms collapsed geographical constraints. Information become real-time and universally accessible.

AI is now accelerating this transformation by compressing innovation cycles, automating cognitive tasks, and enabling small, agile teams to challenge existing companies with an unprecedented leverage.

As a result, mature industries, once defined by predictability, now face rising uncertainty. The ratio of uncertainty to knowledge has increased dramatically. Assumptions that once held for decades may now hold for only a few quarters.

In such an environment, decision-making frameworks built on assumed stability begin to produce fragile outcomes. The more aggressively we attempt to optimize for certainty in one dimension (efficiency, speed, scale), the more variability we often introduce in another (adaptability, resilience, or long-term value creation).

Forecasting accuracy narrows strategic flexibility. Hyper-optimization reduces optionality. Control over one variable amplifies uncertainty over another. And all of these because every system (physical, economic, or organizational) contains trade-offs that cannot be eliminated, only managed.

We live in an era where patterns seem to emerge faster than ever. Market shifts appear visible on dashboards. Adoption curves can be tracked in real time. Competitive moves are amplified instantly. In such an environment, it is tempting to believe that with enough data, enough analytics, and advanced AI models, the future can be predicted with increasing accuracy.

This belief is seductive and I would even say dangerous. It suggests that because change appears directional, the outcomes are foreseeable. It invites extrapolation. It rewards confidence.

While psychologically and humanly we might find some comfort in this, the uncertainty principle whispers a higher truth: the future unfolds in patterns we cannot fully predict. The organizations that rise above it are those that embrace this uncertainty, stepping forward with vision, shaping the currents of change, and turning unpredictability into their greatest opportunity.

II. The illusion of obvious momentum

In our tech world, momentum seems to be everywhere. Markets rally. AI capabilities improve with each release. Startups scale at unprecedented speed. Consumer trends explode overnight. Platforms consolidate power… Entire industries appear to be heading in a clear direction.

From the outside, it looks linear.

Momentum feels visible, almost tangible. Charts trends upward. Adoption accelerates. Headlines declare inevitability. Capital flows decisively. Leaders speak of transformation as if it is already complete.

But… (there’s always a „but”), just because the direction seems obvious doesn’t mean the path is predictable. In fact, it might just be an assumption that acceleration implies sustainability, based simply on the comfort of extrapolating yesterday’s curve into tomorrow’s certainty. Any thoughts on that?

There is no riskier word in strategy than „obvious.”

In moments of rapid technological acceleration, certain trajectories begin to appear clearly dominant. Market share consolidates. Media narratives align. Capital flows consolidate perceived leaders. Executives benchmark against the same platforms. Analysts describe outcomes as inevitable.

Today, in the field of AI, we hear this language again. Certain architectures are described as clearly ahead. Certain vendors are framed structurally dominant. Certain integration models are considered the long-term standard.

However, AI is not simply a fast-moving technology layered onto a stable world, hence the challenge. It is embedded in (and actively reshaping) complex adaptive systems whose structure is constantly evolving. Beneath the surface, the systems driving this momentum are structurally unstable, probabilistic, and deeply interconnected.

If until now, we have learned to survive volatility in stable environments, going through it with clarity and courage. We have come to respect the delicate balance of chaotic systems, understanding that even the smallest change from the beginning can be reflected in extremely different futures. The current structural unpredictability of AI evolution is a different rhythm… AI is a structural variable, not a linear input.

Most historical technologies were additive. They improved processes without fundamentally rewriting the economic grammar of industries. AI is different. It is a general-purpose capability that simultaneously affects decision-making, automation, creativity, communication, and coordination. It alters the labor economy, compressing development cycles, and reshaping competitive barriers.

Even more AI is still in its non-linear phase. Cost curves are constantly changing. Capability ceilings are unclear. Regulatory frameworks are fluid. Public trust is being formed in real time. The architecture of the ecosystem (who integrates with whom, what standards dominate, how value is captured) is not settled, yet.

In such an environment, structural unpredictability means that:

• Minor model improvements can reorder competitive hierarchies.
• Minor changes in capital allocation can accelerate consolidation.
• Interface changes can redefine entire user behavior.
• Policy updates can alter global investment flows.

In other words, the system is not simply evolving, but rather its sensitivities are sharpening. Each adjustment is no longer just a change in outcomes, but a change in the rules that shape those outcomes. Each refinement, each consequence is equally systemic.

For businesses, this invalidates the comfort of any extrapolation. It cannot be assumed that because the system behaves a certain way today, small adjustments will produce proportionally small effects tomorrow. In structurally unstable systems, linear reasoning fails.

Consider these three dimensions:

• In engineering, a system exhibits structural fragility when a minor joint failure leads to disproportionate collapse. In AI ecosystems, a small vulnerability in data governance or model security can trigger a regulatory backlash that reshapes the entire industry’s compliance architecture.
• In economics, a small adjustment in regulations can produce a structural break, not only slowing growth but also redefining incentive structures. In AI markets, a change in intellectual property, privacy enforcement, or export controls could reorganize global competitive dynamics.
• In complexity science, systems are structurally unknowable when the web of interactions is too dense to be perfectly modeled. AI systems, embedded in enterprises, governments, consumer platforms, and global supply chains, create precisely such dense interdependencies. A localized intervention can have negligible impact or trigger a cascading reconfiguration.

We often cannot know in advance what it will be. The rules of the system are still being negotiated (technically, economically and politically). This makes the present moment unique and extremely sensitive.

For businesses, this reframes experimentation:

• A limited pilot can unlock transformative efficiencies.
• A minor weakness in data governance can escalate into a systemic regulatory risk.
• A marginal improvement in user experience can determine market category ownership.

In such systems, momentum is deceptive. It is not a steady forward movement. It is the accumulation of compounding micro-advantages that can suddenly manifest as discontinuity. Businesses that understand this treat each initiative not as incremental progress but as a potential seed of inflection.

Another key point is that AI is not a product category, but an ecosystem. It operates within a network of developers, regulators, enterprises, consumers, hardware providers, cloud platforms, capital markets, media narratives, and geopolitics. No single actor controls its outcome. Each participant shapes the trajectory.

What does this mean? It means that there is system-level behavior that results from countless interactions. When large-scale generative models exceed usability thresholds, the impact is not limited to technology. It triggers enterprise-level experimentation, regulatory debates, capital reallocations, startup creation, organizational restructuring, and labor market anxiety.

For businesses, the implication is profound. You are not reacting to AI, you are participating in shaping it. Your adoption decisions influence competitor behavior. Your workforce strategies alter labor dynamics. Your governance choices signal regulatory expectations. Your customer-facing applications reset user benchmarks.

Momentum in such a system is co-created. The illusion lies in assuming it is externally determined.

The mistake many companies make is asking, „Which AI platform will win?” That’s a deterministic question in a probabilistic world. A better question is, „What range of AI futures is plausible, and how do we position ourselves for them?”

AI development follows distributions, not certainties. Multiple models can coexist. Open-source and proprietary ecosystems can thrive simultaneously. Regulations can accelerate in one geographic area while remaining permissive in another. Capability growth may plateau temporarily, then spike.

In such conditions, strategy must become portfolio-based rather than prediction-based. Resilient organizations avoid exclusive reliance when flexibility is feasible. They invest in internal AI literacy rather than outsourcing understanding. They design modular systems capable of adaptation. They develop governance structures that evolve with regulatory realities rather than hard-coding rigid policies.

The goal is not to predict the winning architecture. But to remain viable across plausible futures. While momentum tempts businesses to overcommit, probabilistic thinking disciplines them to diversify.

Another strong aspect specific to AI markets is that they exhibit strong feedback loops. More users generate more data. More data improves models. Better models attract more users. Enterprise adoption deepens integration. Integration increases switching costs. Switching costs reinforce employment.

This produces rapid concentration. Small early leads compound quickly. Apparent momentum accelerates. However, again, the illusion arises. The pattern (consolidation) is predictable. The identity of the consolidator is not.

Likewise in terms of timing. It becomes delicate…Enter too late, and switching costs rise prohibitively. Enter too early, and standards may shift. Commit too rigidly, and architectural evolution may strand investment.

Momentum in networked markets can reverse faster than it forms. History has repeatedly shown that dominant positions built on a single technological paradigm may erode when the paradigm shifts.

For businesses operating in 2026 and beyond, understanding these illusions (and not mistaking temporary acceleration for structural permanence) is strategic survival.

III. Predictable unpredictability

AI seems chaotic because it is nonlinear, adaptive, human-influenced, global, and rapid. Breakthroughs emerge unexpectedly. Public narratives shift dramatically. Regulatory announcements introduce volatility. Venture capital funding ebbs and flows.

Yet beneath the turbulence, structural trends persist:

• Capacity growth trends upward.
• Cost curves trends downward.
• Automation expands.
• Integration deepens.
• Regulation gradually formalizes.

Organizations leading AI research labs and major technology companies are driving rapid progress. Yet, leadership positions shift, alliances evolve, and breakthroughs often emerge from unexpected quarters.

The pattern may appear stable, while the path is still volatile. This is predictable unpredictability.

However, structural unpredictability does not imply paralysis. It requires a different kind of discipline. It requires scenario awareness beyond incremental forecasts. It requires diversification across platforms and architectures, when feasible. It requires capital buffers to survive structural disruptions. It requires internal learning capacity so that shifts are detected early.

Above all, it requires intellectual humility. Leaders must accept that they are not managing within a fixed system. They are navigating and contributing to a system whose architecture is still being formed.

AI has placed the global economy at a structurally sensitive juncture. Tiny changes in rules, parameters, and connections can reshape entire industries. The shift from prediction to resilience is becoming a structural necessity.

IV. Leadership mindset

Engineers who design bridges in seismic zones do not optimize for average loads, they design for extreme stress. Policymakers who manage financial systems attempt to identify systemic critical points where small shocks can cause cascading failures.

Business leaders in the AI era must adopt a similar thinking:

Where are the structural dependencies in our organization? Which vendors, data sources, or platforms represent single points of failure? What regulatory assumptions underpin our business model? What cost curves need to remain stable for our margins to hold?

More importantly: What would happen if those assumptions shifted abruptly?

When leaders believe the future is clear, they optimize for one scenario. Concentrate capital. Tightly align infrastructure. Eliminate redundancy. Accurately forecast demand. Eliminate resource scarcity.

In stable environments, this maximizes profit. In structurally unpredictable environments (like the one we find ourselves in), overly strategic commitment to betting on a single future could be a fatal error. It only amplifies risk.

A predictive mindset looks for early warning indicators of the next disruption. A resilience mindset accepts that some disruptions will occur without warning and focuses instead on structural preparedness.

Solid balance sheets. Flexible cost structures. Remote-capable operations. Decentralized decision-making. Scenario planning that explores extreme but plausible outcomes. A resilient organization must not see the Black Swan approaching. It must be built in such a way that a massive shock does not lead to existential collapse.

With such vast information flows, it is easy today for predictive models to over-adapt, treating short-term fluctuations as long-term trends. Temporary anomalies can look like structural changes (noise is mistaken for a signal). Strong leaders are aware of this risk, so they do not act decisively on what later turns out to be just a statistical mirage. Especially when the more data is available, the easier it becomes to justify a confident narrative.

Resilience counters this by embracing optionality. Instead of fully committing to a single designed outcome, businesses design portfolios of small, limited experiments. If one fails, the downside is limited. If one succeeds, it can be expanded. Over time, optionality turns into an adaptive advantage.

V. Shaping the pattern

Where prediction concentrates risk, resilience distributes it. The shift from prediction to resilience is not just structural. It is cultural.

In prediction-oriented organizations, failure to anticipate change is punished. Leaders defend outdated plans because deviation implies error. Teams become attached to forecasts, rather than responding to reality.

In resilience-oriented organizations, uncertainty is normalized. Failure is expected within controlled limits. Employees are not penalized for failing to predict the unpredictable, they are encouraged to detect and respond to emerging realities.

Psychological safety increases agility. Agility increases survival. Simple as that… Strategy becomes dynamic, not fixed. Planning becomes scenario-based, not deterministic. Resource allocation incorporates redundancy, instead of pursuing minimal slack at all costs.

The central lesson seems simple, yet profound: You can’t prevent the storm. You can’t predict every shock. Yet you can build a ship that doesn’t need a perfect weather forecast to stay afloat.

***

If you feel more energized than worried…

More curious than defensive…

More interested in shaping outcomes than predicting them…

If you believe momentum is something we generate (not something we chase) then this is a moment to act.

Let’s collaborate.Let’s co-create spaces where probabilistic thinking is operational, not rhetorical. Let’s design experiments instead of waiting for perfect clarity. Let’s build structures resilient enough to survive uncertainty and intelligent enough to evolve through it. Keep it handy!