Engine Non-Goals¶
This document defines what the engine explicitly does not attempt to do.
These are not temporary limitations. They are permanent boundaries.
1. The Engine Does Not Decide What Is "Good"¶
The engine does not:
- rank outputs
- optimise for engagement
- optimise for pedagogy preference
- optimise for market success
- determine "best" explanations
All notions of "better" or "worse" exist outside the engine.
2. The Engine Does Not Possess Taste¶
The engine has no:
- style preference
- tone preference
- ideological stance
- cultural bias
- educational philosophy
Taste belongs to humans and interfaces.
3. The Engine Does Not Invent Intent¶
The engine never:
- creates learning outcomes
- introduces new skills
- reframes educational purpose
- expands scope implicitly
All intent must be declared upstream.
4. The Engine Does Not Personalise by Default¶
The engine does not:
- adapt to individual learners
- infer learner ability
- adjust difficulty implicitly
- "teach differently" unless instructed
Personalisation is an explicit layer above the engine.
5. The Engine Does Not Explain Itself¶
The engine does not:
- justify its decisions
- narrate its reasoning
- expose internal mechanics
- apologise for constraints
Transparency exists through documentation, not behaviour.
6. The Engine Is Not an Educator¶
The engine does not:
- teach
- motivate
- encourage
- persuade
It structures educational material. Teaching happens downstream.
7. The Engine Is Not Responsible for Outcomes¶
The engine does not guarantee:
- learner success
- comprehension
- satisfaction
- compliance with external accreditation
It guarantees structural integrity only.
End of Non-Goals
Version History¶
| Version | Date | Change | Author |
|---|---|---|---|
| 1.0 | 2026-03-11 | Migrated from engine/ucca-engine/docs/engine/ENGINE_NON_GOALS.md | Claude Code |