Automation

Automation in FactHarbor amplifies human capability but never replaces human oversight.
All automated outputs require human review before publication.

This chapter defines:

• What must remain human-only
• What AI (AKEL) can draft
• What can be fully automated
• How automation evolves through POC → Beta 0 → Release 1.0

Manual vs Automated Responsibilities

Human-Only Tasks

These require human judgment, ethics, or contextual interpretation:

• Definition of key terms in claims
• Approval or rejection of scenarios
• Interpretation of evidence in context
• Final verdict approval
• Governance decisions and dispute resolution
• High-risk domain oversight
• Ethical boundary decisions (especially medical, political, psychological)

Semi-Automated (AI Draft → Human Review)

AKEL can draft these, but humans must refine/approve:

• Scenario structures (definitions, assumptions, context)
• Evaluation methods
• Evidence relevance suggestions
• Reliability hints
• Verdict reasoning chains
• Uncertainty and limitations
• Scenario comparison explanations
• Suggestions for merging or splitting scenarios
• Draft public summaries

Fully Automated Structural Tasks

These require no human interpretation:

• Claim normalization
• Duplicate & cluster detection (vector embeddings)
• Evidence metadata extraction
• Basic reliability heuristics
• Contradiction detection
• Re-evaluation triggers
• Batch layout generation (diagrams, summaries)
• Federation integrity checks

Automation Roadmap

Automation increases with maturity.

POC (Low Automation)

Automated

• Claim normalization
• Light scenario templates
• Evidence metadata extraction
• Simple verdict drafts (internal only)

Human

• All scenario definitions
• Evidence interpretation
• Verdict creation
• Governance

Beta 0 (Medium Automation)

Automated

• Detailed scenario drafts
• Evidence reliability scoring
• Cross-scenario comparisons
• Contradiction detection (local + remote nodes)
• Internal Truth Landscape drafts

Human

• Scenario approval
• Final verdict validation

Release 1.0 (High Automation)

Automated

• Full scenario generation (definitions, assumptions, boundaries)
• Evidence relevance scoring and ranking
• Bayesian verdict scoring across scenario sets
• Multi-scenario summary generation
• Anomaly detection across nodes
• AKEL-assisted federated synchronization

Human

• Final approval of all scenarios and verdicts
• Ethical decisions
• Oversight and conflict resolution

Automation Levels

Level 0 — Human-Centric (POC)

AI is purely advisory, nothing auto-published.

Level 1 — Assisted (Beta 0)

AI drafts structures; humans approve each part.

Level 2 — Structured (Release 1.0)

AI produces near-complete drafts; humans refine.

Level 3 — Distributed Intelligence (Future)

Nodes exchange embeddings, contradiction alerts, and scenario templates.
Humans still approve everything.

Automation Matrix

Always Human

• Final verdict approval  
• Scenario validity  
• Ethical decisions  
• Dispute resolution  

Mostly AI (Human Validation Needed)

• Claim normalization  
• Clustering  
• Evidence metadata  
• Reliability heuristics  
• Scenario drafts  
• Contradiction detection  

Mixed

• Definitions of ambiguous terms  
• Boundary choices  
• Assumption evaluation  
• Evidence selection  
• Verdict reasoning  

Diagram References