1. Definition
Reflective Practice for adults is the structural integration of cognitive friction into real-world professional environments, designed to slow down the automatic acceptance of AI outputs in high-pressure contexts.
2. Use Case
Activated in the production of critical professional deliverables (e.g., software architectures, strategic plans, clinical assessments) where AI agents are used to accelerate development or generate first drafts.
3. Human Role
Maintains mature autonomy: defines acceptance criteria before generation, frames the specific domain context, validates the output, and assumes exclusive legal, ethical, and technical responsibility for the final result.
4. AI Role
Functions as a production accelerator and Diagnostic Checker. It operates strictly within the boundaries and constraints established by the professional, providing high-speed syntheses or variations.
5. Friction
Imposes an explicit “accountability checkpoint” in the workflow. The professional cannot approve or implement the generated code/text without first validating it against the pre-established parameters.
6. Risk
The absence of this practice in teams leads to cognitive_debt and automation_bias: professionals trust outputs blindly, gradually losing mastery of the complex systems they manage.
7. Observable Markers
The professional documents (in Git commits or release notes) the exact criteria by which the AI’s output was validated, highlighting which portions were rejected or modified to fit the human context.