Stay Above the Curve

Fear is a signal, not a strategy.

AI tools keep getting better at turning intent into output. That changes the baseline. Execution gets cheaper, and the bar moves up.

The practical question is not, "Will AI replace jobs?"

The question is: will you keep upgrading how work gets done fast enough to stay above the curve?

You are not entitled to the job you got a year ago if you have not upgraded since. That is not moral language, it is unit economics.

In “The Fifth Discipline,” Peter Senge argues for building a learning organization that can keep adapting as the environment shifts [1]. That is the point here. At the leadership level, the job is to make upgrading normal, not heroic. At the individual level, if your work keeps upgrading, you become hard to rationalize removing. If your work stays fixed while the company’s growth curve rises, the gap becomes visible.

The upgrade loop (two weeks)

Most people react to change by working harder.

The higher leverage move is to upgrade a system so the work gets easier, faster, and cleaner every time you do it.

Run this loop every two weeks:

• Pick one workflow that is painful and frequent (reporting, triage, onboarding, hiring, customer response, QA).
• Baseline it with a small scoreboard: time spent, handoffs, defects, latency, rework.
• Upgrade it: simplify steps, remove approvals, standardize inputs, add templates, automate, and use AI where it compresses draft work.
• Add verification: define the check, where evidence shows up, who reviews, and what rollback looks like.
• Ship the new standard: document it, name an owner, and make it easy for others to follow.
• Demo the delta: show the before and after metrics, and what it unlocked downstream.

Upgrades compound. A single workflow upgrade pays you back every week.

Continuous optimization changes the math

Continuous optimization is not a slogan, it is an operating model.

Teams that streamline processes, automate recurring work, and adopt better tools do not just work harder. They change the unit economics of their output.

This aligns with “kaizen,” the Japanese philosophy of continuous improvement [2]. It also matches the “culture of discipline” idea in “Good to Great” [3]. The signal you are sending is simple: we are a compounding asset, not a fixed cost.

Concrete moves (pick a lane and measure it):

• Reduce cycle time for a core workflow by removing handoffs and rework.
• Automate one recurring task per sprint, and track what it unlocks downstream.
• Document the new standard so the improvement sticks (and does not live only in one person’s head).

Exceed the growth curve

AI replacing jobs is creating more anxiety because it is now easier to see the substitution path. Tasks get automated, teams get smaller, and "good enough" output becomes abundant.

In “The Power of Pull,” Hagel and Brown argue that in fast-moving environments, the growth curve required of individuals rises, whether they like it or not [4].

In high-growth companies, performance expectations can rise faster than linear. If you are not upgrading how you work, you can feel busy and still fall behind the curve.

Christensen’s “The Innovator’s Dilemma” is about disruption, but there is a smaller-scale version inside teams [5]. You can disrupt your own workflow before someone else does. You do it by learning, simplifying, automating, and redesigning the role around leverage, not throughput.

The blue dashed line is the growth of the company. The red dashed line is the average growth of the team. Person 05 is below the company’s growth line, which pulls down the team trend. That’s the math underneath the “AI replacing jobs” anxiety.

The move is to make parts of your current work redundant, on purpose. That is how you create capacity to take on the next set of problems, instead of defending the current set.

Run the upgrade loop above, then demo the delta to your manager and your adjacent teams. Repeat until your default contribution is not just output, it is compounding systems improvement.

Tradeoff: you will ship less "new" work in week one. Upside: you ship an improvement that compounds every week after.

What organizations add people for now

AI makes it cheaper to produce drafts. It does not make it cheaper to decide what matters, set constraints, verify quality, and ship changes that hold up in the real world.

When organizations decide who to add, the priority shifts toward people who can:

• Upgrade workflows, not just execute tasks.
• Turn ambiguity into a small bet with a measurable success metric.
• Verify outputs (and catch subtle errors) before they ship.
• Codify improvements so the team benefits, not just the individual.
• Communicate tradeoffs, decision rights, and next steps with clarity.

Weekly cut

Once a week, write three lines:

• Stop: one recurring task you will delete, freeze, or de-optimize.
• Start: one small upgrade you will ship.
• Continue: one behavior that keeps you above the curve.

Stop is the capacity line. If it stays empty, the week will consume whatever slack you have.

Sources

[1] Senge, P. M. (1990). The Fifth Discipline. Doubleday.

[2] Imai, M. (1991). Kaizen: The Key To Japan’s Competitive Success. McGraw-Hill.

[3] Collins, J. (2001). Good to Great. Random House.

[4] Hagel, J., Brown, J. S., & Davison, L. (2012). The Power of Pull. Basic Books.

[5] Christensen, C. M. (2016). The Innovator’s Dilemma. Harvard Business Review Press.