Design for the Job: Role-Specific Task Matrices (T-IPO/LARA)
The strongest implementation evidence: assessment and curricula based on real occupational tasks outperform generic knowledge-based programs. The task matrix (assessment_spec §4) is the tool for that.
1. The problem with generic programs
Definitions emphasizing programming/statistics miss what workers actually need: interpreting outputs, selecting tools, ethical judgment (AI Literacy Assessment Revisited). Generic training produces generic results.
2. The task matrix (assessment_spec §4)
For each role: Role → Task matrix → Assessment result → stored evidence record.
- Define the role (job analysis)
- Task matrix (which AI tasks the role performs)
- Assessed level per task (task-based, objective)
- Store evidence (Art. 4 proof)
3. Role-embedded instruments
- FALCON-AI (role-embedded, CTRL): autonomy, occupation-alignment — CORE task-design template.
- M-SHALF (modular, layered framework): role-specific curricula.
- DGKL: task-oriented assessment.
4. In the ALaaS offer
The task matrix is the bridge between Stage 1 (mini-audit: 4 tasks) and Stage 3 (enterprise: full matrix per role). See ALaaS offer tiers.
Further reading
Sources: Guide 1 §4; assessment_spec §4; corpus of 16,459 papers (as of 2026-10). EOF echo "en7 done"