The Regime of Competence
In The You On AI Encyclopedia
In pre-AI software development, the regime of competence was maintained not by design but by the accumulated complexity of the domain. Debugging sessions, dependency management, cryptic error messages, and the slow translation from human intention to machine instruction kept practitioners continuously stretched. None of this friction was pleasant, but its cumulative effect was to keep developers inside the zone where capability grows. Every four-hour block contained perhaps ten minutes of genuine revelation — moments when the assumption broke, the model revised, the layer deposited. The ten minutes were indistinguishable from the surrounding tedium until they happened.
AI tools dramatically lower the resistance of the environment. The tasks that consumed 80% of a developer's time — Segal's estimate in You On AI — can now be handled by the machine, freeing attention for higher-order work. The liberation is real. But the friction that AI removes is the same friction that maintained the regime. With the challenges fewer, failures less frequent, feedback less specific, and stretch less demanding, the regime thins for a specific class of practitioner: those who begin their careers inside the AI-augmented environment without passing through the older, thicker regime first.
The regime is not unique to software or games. A surgical residency maintains a regime of competence through increasingly complex procedures performed under supervision. A jazz ensemble maintains it through the real-time demand to respond to harmonic surprise. In each case, the regime depends on resistance — on the environment's refusal to collapse to the practitioner's current level. Remove the resistance, and the stretch disappears. Without stretch, there is no learning. The logic is sequential, and each step depends on the one before it.
The practical question is not whether AI should be used but whether organizations and educators can design environments that maintain the regime within AI-augmented workflows. The answer is not to reintroduce arbitrary friction — tedium is not itself educational. It is to preserve the specific kind of friction that produces stretch: the pleasantly frustrating challenges that keep the practitioner working at the edge of capability, with enough scaffolding to succeed through effort and enough resistance that success requires the effort.
Origin
Gee introduced the regime of competence in What Video Games Have to Teach Us About Learning and Literacy (2003), where it appeared as one of thirty-six learning principles he identified in well-designed games. The concept drew on Vygotsky's zone of proximal development and Csikszentmihalyi's flow state, but Gee's contribution was to show that games operationalize these principles with a precision that formal education rarely achieves.
Key Ideas
Calibration, not elimination. The regime is maintained by calibrating challenge to capability, not by removing either.
Stretch precedes learning. Without the stretch that difficulty produces, no layer is deposited.
The regime is environmental. It is a property of the learning environment, not of the learner's attitude or the tool's sophistication.
AI thins the regime silently. The output continues to look excellent as the conditions for mastery erode beneath it.
Deliberate design is required. Environments that preserve the regime in the AI age must be built intentionally; the default is erosion.
Debates & Critiques
Whether the regime of competence that AI creates — a regime of direction rather than implementation — produces understanding of equivalent depth is the central empirical question the AI transition has opened. Directing AI is a genuine skill with its own developmental arc. Whether that arc produces the same embodied intuition that implementation practice produced, or produces a different and perhaps thinner form of mastery, will not be settled by argument. It will be settled by the practitioners now forming inside the new regime, and by what they can do when the AI fails and they must fall back on the understanding they did not have the chance to build.
Further Reading
- James Paul Gee, What Video Games Have to Teach Us About Learning and Literacy (Palgrave Macmillan, 2003, rev. 2007)
- Mihaly Csikszentmihalyi, Flow: The Psychology of Optimal Experience (Harper Perennial, 1990)
- Lev Vygotsky, Mind in Society (Harvard University Press, 1978)
- Anders Ericsson and Robert Pool, Peak: Secrets from the New Science of Expertise (Houghton Mifflin Harcourt, 2016)
- Robert and Elizabeth Bjork, "Making things hard on yourself, but in a good way: Creating desirable difficulties to enhance learning" (Psychology and the Real World, 2011)