Winners and Losers in the AI Era— Six Dividing Lines and How to Cross Them

"Winners and losers" is often used carelessly. Salary, title, SNS follower count. This essay does not use those surface markers but defines "winning" as the structural capacity to live well through the AI era over 10-20 years. Building on the intelligence framework of Vol. 8 and the white-collar survival strategy of Vol. 9, this essay maps six dividing lines. It then offers a concrete five-stage migration strategy for those currently on the "losing" side to move to the "winning" side. The essay is intended not for labeling, but for self-examination and action.

01How "Winning" and "Losing" Are Defined Here

First, the definition this essay uses. What is excluded: salary, title, possessions, SNS follower count, educational background — these are short-term and surface indicators that miss the essence of the AI era.

What is adopted: can you sustain and expand the following on a 10-20 year horizon —

• The essential value of your work: do you have a domain in the AI era where you can say "only I can do this"
• The ability to keep learning: can you update yourself in response to a changing environment
• Bearing judgment and accountability: can you say "I decided this" on important matters
• Relationships: a network of trust, dialogue, and collaboration
• Participation in creation: not only solving existing problems but generating new questions
• An ethical compass: judging by your own values rather than being swept along

Those who satisfy these six are "winners" in this essay; those for whom these are slipping are "losers." A person with a high salary who keeps mechanically performing routine tasks AI will replace is — a candidate for the "losing" side in five years. A person with a moderate salary who steadily accumulates the six above is on the "winning" side.

02Dividing Line ① — Do You Have the Posture to Keep Learning

The AI era is one in which the "way of working" is fundamentally rewritten every three years. ChatGPT in 2022, GPT-4/Claude 3 in 2023, reasoning models (o1/o3) and AI agents in 2024, Agentic AI and video generation (Sora/Veo) in 2025, ... in 2026.

Winning side: when a new technology appears, they try it. They spend three months integrating it into their own work. They share with colleagues. They treat "learning the change itself" as part of the job.

Losing side: "I'm too old for this," "what I do now is sufficient," "I'm too busy to learn new things." After three years, their working method is out of date. After five years, juniors think of them as obsolete.

03Dividing Line ② — Can You Design Questions to AI

The gap in AI-era productivity is determined by "what you ask AI" and "how you ask". With the same AI, the quality of the question yields tenfold differences in output.

Winning side's questioning: provide context ("I work in pharmaceutical promotional material review, situation X"), state the objective clearly ("I want to judge Y"), present constraints ("subject to regulation Z"), compare across multiple AIs, ask for sources, continue the dialogue.

Losing side's questioning: ends with "tell me about X." Believes the answer that comes back as-is. Does not check sources. Does not verify with another AI. Gives up after one round, declaring "AI is useless."

This is the difference in proficiency across the eight components shown in Vol. 8. Whether you see AI as a "magic answer machine" or as a "dialogue partner" determines whether your productivity in five years differs by an order of magnitude.

04Dividing Line ③ — Will You Bear Judgment and Accountability

AI can produce answers but cannot take responsibility. The role of "deciding and accepting the consequence" remains in the human domain.

Winning side: holds a domain where they can say "I decided this." Makes calls in gray zones. Responds in their own name when something fails. Does not flee from hard decisions.

Losing side: "Because management said so," "because it's the rule," "because AI said so" — outsourcing the locus of judgment. Blames others when things fail. Avoids gray zones and selects only work with clear black-and-white answers.

05Dividing Line ④ — Can You Build Interpersonal Trust

Precisely because AI can also be a conversational counterpart, the value of "real human relationships" paradoxically rises.

Winning side: builds a savings account of trust with colleagues, clients, juniors, seniors. Has people who help when things get hard. Is told "I want to work with you" beyond transactional considerations. Values long-term relationships.

Losing side: builds only superficial relationships. Has many SNS followers but no one to call for a serious conversation. Relationships die when the immediate transaction ends. Avoids the effort of interpersonal work with "I'm too busy" / "too much hassle."

In an era when AI provides instant answers, "spending three hours over drinks," "deepening a relationship over six months" — these "inefficient" acts become increasingly precious investments. Those who dismiss them as wasteful will find, in five years, that they have no one to rely on.

06Dividing Line ⑤ — Will You Pose Questions and Create

AI answers given questions, but posing the genuinely important question still remains in human hands.

Winning side: "asks what no one has asked," "tries new combinations," "tackles unsolved problems." Stands on the side of business design, research themes, organizational reform, institutional design — generating something new.

Losing side: processes given work as instructed. Does not pose new questions. Does not ask "is this right? are there other approaches?" Lives in "reproduction" rather than creation.

The easiest labor to replace with AI is reproductive labor. Standing on the side of creation is what sustains structural "winning."

07Dividing Line ⑥ — Do You Have an Ethical Compass

Precisely because AI generates an overwhelming abundance of information, options, and recommendations, "what you select and what you refuse" — the ethical compass — becomes decisively important.

Winning side: holds their own values. When short-term gain conflicts with long-term trust, chooses the long term. Has the courage to refuse ethically gray work. Is not swept along.

Losing side: ambiguous values. Prioritizes short-term gain. Outsources ethical judgment with "everyone is doing it / it pays / management said so." Is swept along.

The AI era is one of an exponential increase in options. Those without an ethical compass drown in the sea of options. Those with a compass pick up only the options they need and move at their own pace.

08The Five-Stage Migration Strategy From Losing to Winning

If reading this leaves you thinking "I'm closer to the losing side," there is no need to despair. The migration can happen consciously and incrementally.

Stage 1: Self-diagnosis (this week)

For each of the six dividing lines, score yourself honestly on a 1-5 scale. Have family or a trusted colleague score you too. Pick the one with the largest gap as the first target for improvement.

Stage 2: Small experiments (this month)

For the one dividing line you chose, start small experiments. Example: for "keep learning," try one new Claude or ChatGPT feature each week and share the result with a friend. For "interpersonal trust," go to lunch twice a month with people unrelated to immediate business. Pick reliably doable small actions, not vast goals.

Stage 3: Build a peer group (3 months)

Find 3-5 peers with the same concerns. Inside your company, study groups, SNS, professional society. Even monthly, create a place to say "what did you learn?" / "how are you approaching this?" Sustaining solo effort is hard; sustaining it with peers is possible.

Stage 4: Embed in work (6 months)

Embed the individual experiments into your work process. Example: always include "AI tools learned this month and their applied results" in your monthly status report. Always include one "creative question" session in each project. Change the organization.

Stage 5: Extend to the remaining five (1-3 years)

Once you have a success experience on the first one, extend in sequence to the remaining five dividing lines. 2-3 per year, all six lifted in three years. The conviction "in three years, I can change" emerges.

"Winners and losers" is not innate ability, luck, or current position. It is the accumulation of conscious choices and small daily actions. The six dividing lines are matters of habit and awareness, not personality or age. They can be changed starting today.

The AI era is one in which the gap between effort and stagnation compounds faster than in any prior era. Those who learn keep learning; those who create keep creating; those who build trust keep being trusted. Conversely, those who stop are increasingly left behind. The bifurcation of winners and losers proceeds faster and wider than in the past.

But this is not determinism. The moment you notice, correction is possible. May this essay be one input for self-examination and the next step. What I wanted to convey across this three-essay sequence (Vols. 8-10) is this: do not treat AI as a "threat" or a "magic" but as "a mirror through which to refine yourself as a human being." Whether to stand before that mirror and face it is, in the end, your own choice.