Wiki source code of Automation
Version 5.1 by Robert Schaub on 2025/12/12 15:41
Hide last authors
| author | version | line-number | content |
|---|---|---|---|
 |
1.1 | 1 | = Automation = |
| 2 | |||
| 3 | Automation in FactHarbor amplifies human capability but never replaces human oversight. | ||
| 4 | All automated outputs require human review before publication. | ||
| 5 | |||
| 6 | This chapter defines: | ||
| 7 | * What must remain human-only | ||
| 8 | * What AI (AKEL) can draft | ||
| 9 | * What can be fully automated | ||
| 10 | * How automation evolves through POC → Beta 0 → Release 1.0 | ||
| 11 | |||
| |
3.1 | 12 | == POC v1 (Fully Automated "Text to Truth Landscape") == |
| 13 | |||
| 14 | The goal of POC v1 is to validate the automated reasoning capabilities of the data model without human intervention. | ||
| 15 | |||
| 16 | === Workflow === | ||
| 17 | |||
| 18 | 1. **Input**: User pastes a block of raw text. | ||
| |
4.1 | 19 | 1. **Deep Analysis (Background)**: The system autonomously performs the full pipeline **before** displaying the text: |
| 20 | * Extraction & Normalisation | ||
| 21 | * Scenario & Sub-query generation | ||
| 22 | * Evidence retrieval & Verdict computation | ||
| 23 | 1. **Visualisation (Extraction & Marking)**: The system displays the text with claims extracted and marked. | ||
| 24 | * **Verdict-Based Coloring**: The extraction highlights (e.g. Orange/Green) are chosen **according to the computed verdict** for each claim. | ||
| 25 | 1. **Inspection**: User clicks a highlighted claim to see the **Reasoning Trail**, showing exactly which evidence and sub-queries led to that verdict. | ||
| |
3.1 | 26 | |
| 27 | === Technical Scope === | ||
| 28 | |||
| 29 | * **Fully Automated**: No human-in-the-loop for this phase. | ||
| 30 | * **Structured Sub-Queries**: Logic is generated by decomposing claims into the FactHarbor data model. | ||
| 31 | * **Latency**: Focus on accuracy of reasoning over real-time speed for v1. | ||
| 32 | |||
| |
1.1 | 33 | ---- |
| 34 | |||
| |
5.1 | 35 | = Manual vs Automated Responsibilities = |
| |
1.1 | 36 | |
| |
5.1 | 37 | == Human-Only Tasks == |
| |
1.1 | 38 | |
| 39 | These require human judgment, ethics, or contextual interpretation: | ||
| 40 | |||
| 41 | * Definition of key terms in claims | ||
| 42 | * Approval or rejection of scenarios | ||
| 43 | * Interpretation of evidence in context | ||
| 44 | * Final verdict approval | ||
| 45 | * Governance decisions and dispute resolution | ||
| 46 | * High-risk domain oversight | ||
| 47 | * Ethical boundary decisions (especially medical, political, psychological) | ||
| 48 | |||
| |
5.1 | 49 | == Semi-Automated (AI Draft → Human Review) == |
| |
1.1 | 50 | |
| 51 | AKEL can draft these, but humans must refine/approve: | ||
| 52 | |||
| 53 | * Scenario structures (definitions, assumptions, context) | ||
| 54 | * Evaluation methods | ||
| 55 | * Evidence relevance suggestions | ||
| 56 | * Reliability hints | ||
| 57 | * Verdict reasoning chains | ||
| 58 | * Uncertainty and limitations | ||
| 59 | * Scenario comparison explanations | ||
| 60 | * Suggestions for merging or splitting scenarios | ||
| 61 | * Draft public summaries | ||
| 62 | |||
| |
5.1 | 63 | == Fully Automated Structural Tasks == |
| |
1.1 | 64 | |
| 65 | These require no human interpretation: | ||
| 66 | |||
| 67 | * Claim normalization | ||
| 68 | * Duplicate & cluster detection (vector embeddings) | ||
| 69 | * Evidence metadata extraction | ||
| 70 | * Basic reliability heuristics | ||
| 71 | * Contradiction detection | ||
| 72 | * Re-evaluation triggers | ||
| 73 | * Batch layout generation (diagrams, summaries) | ||
| 74 | * Federation integrity checks | ||
| 75 | |||
| |
5.1 | 76 | ---- |
| |
1.1 | 77 | |
| |
5.1 | 78 | = Automation Roadmap = |
| 79 | |||
| |
1.1 | 80 | Automation increases with maturity. |
| 81 | |||
| |
5.1 | 82 | == POC (Low Automation) == |
| |
1.1 | 83 | |
| |
5.1 | 84 | === Automated === |
| 85 | * Claim normalization | ||
| 86 | * Light scenario templates | ||
| 87 | * Evidence metadata extraction | ||
| 88 | * Simple verdict drafts (internal only) | ||
| |
1.1 | 89 | |
| |
5.1 | 90 | === Human === |
| 91 | * All scenario definitions | ||
| 92 | * Evidence interpretation | ||
| 93 | * Verdict creation | ||
| 94 | * Governance | ||
| |
1.1 | 95 | |
| |
5.1 | 96 | == Beta 0 (Medium Automation) == |
| |
1.1 | 97 | |
| |
5.1 | 98 | === Automated === |
| 99 | * Detailed scenario drafts | ||
| 100 | * Evidence reliability scoring | ||
| 101 | * Cross-scenario comparisons | ||
| 102 | * Contradiction detection (local + remote nodes) | ||
| 103 | * Internal Truth Landscape drafts | ||
| |
1.1 | 104 | |
| |
5.1 | 105 | === Human === |
| 106 | * Scenario approval | ||
| 107 | * Final verdict validation | ||
| |
1.1 | 108 | |
| |
5.1 | 109 | == Release 1.0 (High Automation) == |
| |
1.1 | 110 | |
| |
5.1 | 111 | === Automated === |
| 112 | * Full scenario generation (definitions, assumptions, boundaries) | ||
| 113 | * Evidence relevance scoring and ranking | ||
| 114 | * Bayesian verdict scoring across scenario sets | ||
| 115 | * Multi-scenario summary generation | ||
| 116 | * Anomaly detection across nodes | ||
| 117 | * AKEL-assisted federated synchronization | ||
| |
1.1 | 118 | |
| |
5.1 | 119 | === Human === |
| 120 | * Final approval of all scenarios and verdicts | ||
| 121 | * Ethical decisions | ||
| 122 | * Oversight and conflict resolution | ||
| 123 | |||
| 124 | ---- | ||
| 125 | |||
| 126 | = Automation Levels = | ||
| 127 | |||
| 128 | == Level 0 — Human-Centric (POC) == | ||
| 129 | AI is purely advisory, nothing auto-published. | ||
| 130 | |||
| 131 | == Level 1 — Assisted (Beta 0) == | ||
| 132 | AI drafts structures; humans approve each part. | ||
| 133 | |||
| 134 | == Level 2 — Structured (Release 1.0) == | ||
| 135 | AI produces near-complete drafts; humans refine. | ||
| 136 | |||
| 137 | == Level 3 — Distributed Intelligence (Future) == | ||
| 138 | Nodes exchange embeddings, contradiction alerts, and scenario templates. | ||
| 139 | Humans still approve everything. | ||
| 140 | |||
| 141 | ---- | ||
| 142 | |||
| 143 | = Automation Matrix = | ||
| 144 | |||
| 145 | == Always Human == | ||
| 146 | * Final verdict approval | ||
| 147 | * Scenario validity | ||
| 148 | * Ethical decisions | ||
| 149 | * Dispute resolution | ||
| 150 | |||
| 151 | == Mostly AI (Human Validation Needed) == | ||
| 152 | * Claim normalization | ||
| 153 | * Clustering | ||
| 154 | * Evidence metadata | ||
| 155 | * Reliability heuristics | ||
| 156 | * Scenario drafts | ||
| 157 | * Contradiction detection | ||
| 158 | |||
| 159 | == Mixed == | ||
| 160 | * Definitions of ambiguous terms | ||
| 161 | * Boundary choices | ||
| 162 | * Assumption evaluation | ||
| 163 | * Evidence selection | ||
| 164 | * Verdict reasoning | ||
| 165 | |||
| 166 | ---- | ||
| 167 | |||
| 168 | = Diagram References = | ||
| 169 | |||
| |
2.1 | 170 | {{include reference="FactHarbor.Specification.Diagrams.Automation Roadmap.WebHome"/}} |
| |
1.1 | 171 | |
| |
2.1 | 172 | {{include reference="FactHarbor.Specification.Diagrams.Automation Level.WebHome"/}} |
| |
1.1 | 173 | |
| |
2.1 | 174 | {{include reference="FactHarbor.Specification.Diagrams.Manual vs Automated matrix.WebHome"/}} |