The Essence of Perspective
— How AI Compresses the Management Time Horizon

This essay does not refer to any specific firm or individual. It is a structural reflection on the relationship between management and technology.

1. The problem — "fast" does not mean "far"

AI's role in management expands every year. McKinsey Global Institute (2023) estimated that generative AI could produce $2.6–4.4 trillion in annual economic value worldwide. 72% of CEOs now classify AI as a "strategic priority" (PwC CEO Survey, 2024).

Yet a structural question deserves attention.

Is making management decisions faster the same thing as seeing further?

This essay argues that it is not. AI adoption accelerates the speed of management decision-making while structurally compressing its time horizon. This compression extends to the advisory and consulting functions that historically existed to "see beyond the frame," causing "long-term perspective" to quietly disappear from management.

2. The mechanics of time-horizon compression

2-1. The temporal bias of data

AI-assisted decision-making depends on training data and input data. These are, by nature, records of the past and present — they do not contain "a future that does not yet exist." Consequently, AI-supported decisions are structurally biased toward optimizing for extensions of the present (Marcus & Davis, 2019).

Clayton Christensen (1997) identified a structural blind spot in The Innovator's Dilemma: existing evaluation metrics cannot detect the emergence of disruptive innovation. AI has dramatically improved the precision and speed of judgments based on existing metrics. But it has no capacity to question the validity of the metrics themselves.

"Being rational and being right are different things. Rationality only functions within an existing framework."

— Herbert Simon, Administrative Behavior (1947)

2-2. Sprint shortening and the convergence of vision

AI-driven acceleration shortens project cycles. The 2–4 week sprints of agile development have become the norm, and this "short-cycle iteration" mentality has permeated management as a whole.

The issue is not shorter cycles per se. It is that only metrics measurable within short cycles get treated as "strategy," while thinking about long-range structural change is dismissed as "inefficient."

• In Kahneman's (2011) framework, management decision-making is skewing toward System 1 (fast, intuitive) while System 2 (deliberate, long-range) atrophies organizationally
• Senge (1990) warned in The Fifth Discipline of "event-level reactivity that blinds organizations to structural change" — AI is accelerating exactly this
• From Taleb's (2012) antifragility lens, the pursuit of short-term optimization creates a paradox: it increases systemic fragility

2-3. The self-reinforcing loop of "short-term rationality"

Most critically, this time-horizon compression is accepted not as irrational but as rational.

AI-generated analysis is quantitative, explainable, and reproducible. "This quarter's revenue forecast" is precise; "the market structure in 10 years" is not computable. Decisions therefore concentrate on the former, and the latter is shelved as "too uncertain."

This is a form of framing effect. What is measurable appears important; what is unmeasurable is treated as if it does not exist. By raising the resolution of the measurable domain to its limit, AI is pushing the unmeasurable domain — long-term structural change — out of the field of vision.

3. The structural transformation of advisory functions

3-1. From "asking questions" to "selling means"

Why have management advisors, consultants, and strategic counselors existed throughout history? Peter Drucker (1954) defined the consultant's value as "asking the questions that cannot be seen from inside the organization."

The critical word is not "answers" but "questions."

• "Will this business exist in 10 years?"
• "How will this decision affect the next generation?"
• "What structural shift is invisible right now?"

These questions share a common trait: they transcend the current time horizon and cannot be measured by existing KPIs. What AI can replace is the "means" side — data analysis, market research, competitive benchmarking. As this substitution progresses, consulting itself is being reduced to "delivery of means," while its essence — "delivery of perspective" — fades.

3-2. Competitive pressure on advisory firms

Consulting firms themselves are exposed to AI-era competitive dynamics. As clients acquire their own AI-powered analytical capabilities, the "selling analysis" model becomes harder to sustain. Firms consequently shift toward shorter-term, implementation-focused services, and "asking about the next decade" type engagements exit the market.

This is not specific to consulting. It is a structural problem: the "long-range vision function" of management is being contracted by market forces.

4. Empirical evidence — how time horizons change decision quality

4-1. Corporate time horizons

The average lifespan of a Japanese company is 23.3 years (Tokyo Shoko Research, 2023). Yet over 33,000 Japanese firms have survived more than 100 years (Teikoku Databank, 2022). Japan has the world's highest count of century-old companies, with 1,340 firms exceeding 200 years.

What these long-lived firms share is not technology or capital but a question at the center of management: "What do we pass to the next generation?" A Kyoto confectioner who thinks about "the flavor 30 years from now" while preparing today's batch operates from a structurally different decision framework.

De Geus (1997) identified in The Living Company that a common trait of long-lived companies is "identity of purpose taking precedence over profit." When existential purpose — not short-term revenue — drives management, corporate lifespan extends structurally.

4-2. The time horizon of memory

Personal memory obeys the same structure. The average cloud service lifespan is 10–15 years. 99% of social media posts go unaccessed after 5 years (Internet Archive, 2021). Digital is an extraordinary tool optimized for "now," but it is not designed to endure for a century.

In Japan, tens of thousands of gravestones are removed as "unclaimed" each year. Stones erected 100 years ago with the wish of "eternity" are vanishing under institutional and environmental change. This illustrates the structural limit of any recording medium designed on a short time horizon.

The question "What would a record designed to last 1,000 years look like?" does not arise from inside a 6-month sprint. It does not register as a valid question from inside a quarterly KPI cycle. This is not a matter of ability. It is a matter of perspective.

5. AI as liberation — possibilities and conditions

This essay is not an argument against AI. It is the opposite.

If AI assumes day-to-day analysis, prediction, and automation, then humans should be free to look further. If AI manages this quarter's numbers, executives can focus on structural change over the next decade. If market analysis is automated, consultants can dedicate themselves to seeing "outside the market."

"The best way to predict the future is to invent it."

— Alan Kay, Xerox PARC (1971)

If Kay's proposition holds, then AI should not be a tool for "predicting" the future but one that gives humans the margin to "invent" it. Yet reality has moved in the opposite direction. AI spins the short-term cycle faster, and humans get caught in the spin. The tool is dictating the human time horizon.

Conditions for liberation

For AI to "liberate" rather than "compress" perspective, the following structural conditions are required:

1. Intentional separation of time horizons — Delegate short-term optimization to AI; concentrate human decision-making resources on 10- to 100-year structural shifts
2. Investment in unmeasurable questions — Allocate organizational resources to "existential purpose," "intergenerational stewardship," and "cultural value" — things no KPI can capture
3. Redefinition of advisory functions — From "means provider" to "time-horizon expander." Evaluate consultants by the quality of their questions, not the speed of their deliverables

6. Conclusion — perspective is structure, and it is a choice

Perspective is not talent. It is structure, and it is a choice.

Which time horizon you stand on can be designed — organizationally and individually. The quarter. The decade. The century. That choice determines everything you see.

If AI is compressing management's time horizon, it is not a flaw of AI. It is a choice made by humans who decided to operate AI only inside the frame. If consulting has been reduced to trading in means, it is a consequence of market structures that reward short-term margin over long-range perspective.

And if memory is disappearing, it is not a limitation of digital. It is the result of a design philosophy that set "present convenience" as its only objective and never asked about a hundred years from now.

"The truly important questions are always outside the current time horizon."

The essence of perspective is which time horizon you stand on when you look at the world — nothing more. And that time horizon can be compressed by technology or expanded by will.

What is being asked right now is not about the evolution of technology. It is about how far the humans using it are willing to look.