The Last Skill?
Finding Our Humanity at the Edge of AI Capability
When was the last time you felt truly inspired?
By a class, a product, a quiet act of kindness—or by the riotous beauty that sometimes follows chaos, when an individual or an organization slowly rewires its own thinking.
Facts and anecdotes can replay the surface of that moment.
But they rarely reveal the deeper story:
How do you actually learn?
Why do you learn the way you do?
Learning is not a file transfer.
It is full-body, full-mind, full-heart (Kolb, 1984).
It rides on friction, failure, curiosity, recovery—the messy processes often hidden even from the learner themselves.
When friction disappears, so does growth.
Recent lab work shows the same effect at scale: learners who rely on LLM summaries form shallower, less original knowledge than those who actively search and synthesize information themselves (Melumad & Yun, 2025).
In an era of computational intelligence—where answers arrive pre-packaged and effort feels optional—this messy, human pathway matters more than ever.
Training vs. Education
A line from a military readiness manual keeps echoing in my mind:
“We are trained for certainty and educated for uncertainty”
-General Schoomaker, 1998 congressional testimony
Training equips us for what we can predict.
Education prepares us for what we cannot (Dewey, 1938).
Today, when almost nothing stays certain for long, we must swing back—from narrow task-training toward true education, from stuffing facts in to drawing wisdom out.
(The Latin educere means “to lead out.”)
The Goldilocks Zone of Growth
Real inspiration isn’t about content transfer. It’s about capacity awakening.
It places us in the narrow “just-right” band where challenge meets safety (Csikszentmihalyi, 1990):
Growth demands friction (McEwen, 2007).
Strength demands movement (Larsson et al., 2019).
Order demands surviving disorder (Taleb, 2012).
Wisdom demands lived experience (Baldwin, 2019).
And depth demands deliberate synthesis of raw sources (Melumad & Yun, 2025)
Agency, Autonomy, and the AI Divide
Artificial agents can already adapt to uncertainty—updating weights, rewriting code.
Humans, however, are transformed by uncertainty.
We experience risk, responsibility, and meaning.
We reorganize ourselves from the inside out.
That transformation is rooted in agency and autonomy—the felt sense of “I can act” (Hitlin & Elder, 2007; Ryan & Deci, 2020).
No machine, no matter how embodied or advanced, has crossed that phenomenological line (Haddadin, 2025; Lake, Ullman, Tenenbaum, & Gershman, 2017).
AI adapts to uncertainty.
Humans become through it.
The difference isn’t in the reaction—it’s in the becoming.
The Human Safeguard
Algorithms can predict, remix, and even simulate inspiration.
They cannot—not yet—fake inner transformation. Even their greatest trick—instant synthesis—can stunt ours, unless we re-insert friction on purpose.
The next time you need to learn something, let the AI spit out its summary—but choose to chase three original sources and write your own synthesis. Then compare the results.
Re-learning how to learn, re-learning how to be inspired, isn’t a luxury.
It’s the last safeguard of human potential.
It’s what keeps you—and all of us—human.
References and Further Exploration
Baldwin, C. K. (2019). Transformative learning and identity: A review and synthesis of Dirkx and Illeris. Adult Education Research Conference, 4149. Kansas State University Libraries New Prairie Press.
Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience. Harper & Row.
Dewey, J. (1938). Experience and education. Kappa Delta Pi.
Haddadin, S. (2025, February 11). AI that moves, adapts, and learns: The future of embodied intelligence. Columbia AI.
Hitlin, S., & Elder, G. H. (2007). Time, self, and the curiously abstract concept of agency. Sociological Theory, 25(2), 170–191.
Kolb, D. A. (1984). Experiential learning: Experience as the source of learning and development. Prentice-Hall.
Lake, B. M., Ullman, T. D., Tenenbaum, J. B., & Gershman, S. J. (2017). Building machines that learn and think like people. Behavioral and Brain Sciences, 40, e253.
Larsson, L., Degens, H., Li, M., Salviati, L., Lee, Y. I., Thompson, W., Kirkland, J. L., & Sandri, M. (2019). Sarcopenia: Aging-related loss of muscle mass and function. Physiological Reviews, 99(1), 427–511.
McEwen, B. S. (2007). Physiology and neurobiology of stress and adaptation: central role of the brain. Physiological Reviews, 87(3), 873–904.
Melumad, S., & Yun, J. H. (2025). Experimental evidence of the effects of large language models versus web search on depth of learning. SSRN Working Paper Series.
Ryan, R. M., & Deci, E. L. (2020). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary Educational Psychology, 25(1), 54–67.
Taleb, N. N. (2012). Antifragile: Things that gain from disorder. Random House.
Schoomaker, P. J. (1998). Congressional testimony on special-operations education. U.S. House Committee on Appropriations.
I’ve been following Dr. Michael J. Jabbour since his standout AI presentation at the 2024 AI and the Future of Education Conference. His insights on Substack are sharp, relevant, and a must-read for anyone exploring Artificial Intelligence.