Changes for page Automation

Summary

• Page properties (2 modified, 0 added, 0 removed)

Details

Page properties

Parent

...	...	@@ -1,1 +1,1 @@
1		-FactHarbor.Archive.FactHarbor.Specification.WebHome
	1	+FactHarbor.Archive.FactHarbor V0\.9\.18.Specification.WebHome

Content

...	...	@@ -3,6 +3,7 @@
3	3	Automation in FactHarbor amplifies human capability while implementing risk-based oversight.
4	4
5	5	This chapter defines:
	6	+
6	6	* Risk-based publication model
7	7	* Quality gates for AI-generated content
8	8	* What must remain human-only
...	...	@@ -18,14 +18,18 @@
18	18
19	19	1. **Input**: User pastes a block of raw text.
20	20	1. **Deep Analysis (Background)**: The system autonomously performs the full pipeline **before** displaying the text:
	22	+
21	21	* Extraction & Normalisation
22	22	* Scenario & Sub-query generation
23	23	* Evidence retrieval with **contradiction search**
24	24	* Quality gate validation
25	25	* Verdict computation
	28	+
26	26	1. **Visualisation (Extraction & Marking)**: The system displays the text with claims extracted and marked.
	30	+
27	27	* **Verdict-Based Coloring**: The extraction highlights (e.g. Orange/Green) are chosen **according to the computed verdict** for each claim.
28	28	* **AI-Generated Label**: Clear indication that content is AI-produced
	33	+
29	29	1. **Inspection**: User clicks a highlighted claim to see the **Reasoning Trail**, showing exactly which evidence and sub-queries led to that verdict.
30	30
31	31	=== Technical Scope ===
...	...	@@ -44,11 +44,13 @@
44	44	FactHarbor implements a risk-based publication model with three modes:
45	45
46	46	=== Mode 1: Draft-Only ===
	52	+
47	47	* Failed quality gates
48	48	* High-risk content pending expert review
49	49	* Internal review queue only
50	50
51	51	=== Mode 2: AI-Generated (Public) ===
	58	+
52	52	* Passed all quality gates
53	53	* Risk tier B or C
54	54	* Clear AI-generated labeling
...	...	@@ -55,6 +55,7 @@
55	55	* Users can request human review
56	56
57	57	=== Mode 3: Human-Reviewed ===
	65	+
58	58	* Validated by human reviewers/experts
59	59	* "Human-Reviewed" status badge
60	60	* Required for Tier A content publication
...	...	@@ -66,6 +66,7 @@
66	66	== Risk Tiers and Automation Levels ==
67	67
68	68	=== Tier A (High Risk) ===
	77	+
69	69	* **Domains**: Medical, legal, elections, safety, security
70	70	* **Automation**: AI can draft, human review required for "Human-Reviewed" status
71	71	* **AI publication**: Allowed with prominent disclaimers and warnings
...	...	@@ -72,6 +72,7 @@
72	72	* **Audit rate**: Recommendation: 30-50%
73	73
74	74	=== Tier B (Medium Risk) ===
	84	+
75	75	* **Domains**: Complex policy, science, causality claims
76	76	* **Automation**: AI can draft and publish (Mode 2)
77	77	* **Human review**: Optional, audit-based
...	...	@@ -78,6 +78,7 @@
78	78	* **Audit rate**: Recommendation: 10-20%
79	79
80	80	=== Tier C (Low Risk) ===
	91	+
81	81	* **Domains**: Definitions, established facts, historical data
82	82	* **Automation**: AI publication default
83	83	* **Human review**: On request or via sampling
...	...	@@ -112,6 +112,7 @@
112	112	* **Public summaries** and accessibility text
113	113
114	114	Most Tier B and C content remains in AI-draft status unless:
	126	+
115	115	* Users request human review
116	116	* Audits identify errors
117	117	* High engagement triggers review
...	...	@@ -139,26 +139,30 @@
139	139	Before AI-draft publication (Mode 2), content must pass:
140	140
141	141	1. **Source Quality Gate**
142		- * Primary sources verified
143		- * Citations complete and accessible
144		- * Source reliability scored
145	145
	155	+* Primary sources verified
	156	+* Citations complete and accessible
	157	+* Source reliability scored
	158	+
146	146	2. **Contradiction Search Gate** (MANDATORY)
147		- * Counter-evidence actively sought
148		- * Reservations and limitations identified
149		- * Bubble detection (echo chambers, conspiracy theories)
150		- * Diverse perspective verification
151	151
	161	+* Counter-evidence actively sought
	162	+* Reservations and limitations identified
	163	+* Bubble detection (echo chambers, conspiracy theories)
	164	+* Diverse perspective verification
	165	+
152	152	3. **Uncertainty Quantification Gate**
153		- * Confidence scores calculated
154		- * Limitations stated
155		- * Data gaps disclosed
156	156
	168	+* Confidence scores calculated
	169	+* Limitations stated
	170	+* Data gaps disclosed
	171	+
157	157	4. **Structural Integrity Gate**
158		- * No hallucinations detected
159		- * Logic chain valid
160		- * References verifiable
161	161
	174	+* No hallucinations detected
	175	+* Logic chain valid
	176	+* References verifiable
	177	+
162	162	See [[AKEL page>>FactHarbor.Specification.AI Knowledge Extraction Layer (AKEL).WebHome]] for detailed quality gate specifications.
163	163
164	164	----
...	...	@@ -168,6 +168,7 @@
168	168	Instead of reviewing all AI output, systematic sampling audits ensure quality:
169	169
170	170	=== Stratified Sampling ===
	187	+
171	171	* Risk tier (A > B > C sampling rates)
172	172	* Confidence scores (low confidence → more audits)
173	173	* Traffic/engagement (popular content audited more)
...	...	@@ -175,7 +175,9 @@
175	175	* User flags and disagreement signals
176	176
177	177	=== Continuous Improvement Loop ===
	195	+
178	178	Audit findings improve:
	197	+
179	179	* Query templates
180	180	* Source reliability weights
181	181	* Contradiction detection algorithms
...	...	@@ -183,6 +183,7 @@
183	183	* Bubble detection heuristics
184	184
185	185	=== Transparency ===
	205	+
186	186	* Audit statistics published
187	187	* Accuracy rates by tier reported
188	188	* System improvements documented
...	...	@@ -196,6 +196,7 @@
196	196	=== POC (Current Focus) ===
197	197
198	198	**Automated:**
	219	+
199	199	* Claim normalization
200	200	* Scenario template generation
201	201	* Evidence metadata extraction
...	...	@@ -205,6 +205,7 @@
205	205	* **Risk tier assignment**
206	206
207	207	**Human:**
	229	+
208	208	* High-risk content validation (Tier A)
209	209	* Sampling audits across all tiers
210	210	* Quality standard refinement
...	...	@@ -213,6 +213,7 @@
213	213	=== Beta 0 (Enhanced Automation) ===
214	214
215	215	**Automated:**
	238	+
216	216	* Detailed scenario generation
217	217	* Advanced evidence reliability scoring
218	218	* Cross-scenario comparisons
...	...	@@ -221,6 +221,7 @@
221	221	* **Increased AI-draft coverage** (more Tier B content)
222	222
223	223	**Human:**
	247	+
224	224	* Tier A final approval
225	225	* Audit sampling (continued)
226	226	* Expert validation of complex domains
...	...	@@ -229,6 +229,7 @@
229	229	=== Release 1.0 (High Automation) ===
230	230
231	231	**Automated:**
	256	+
232	232	* Full scenario generation (comprehensive)
233	233	* Bayesian verdict scoring across scenarios
234	234	* Multi-scenario summary generation
...	...	@@ -237,6 +237,7 @@
237	237	* **Most Tier B and all Tier C** auto-published
238	238
239	239	**Human:**
	265	+
240	240	* Tier A oversight (still required)
241	241	* Strategic audits (lower sampling rates, higher value)
242	242	* Ethical decisions and policy
...	...	@@ -246,13 +246,13 @@
246	246
247	247	== Automation Levels Diagram ==
248	248
249		-{{include reference="Test.FactHarborV09.Specification.Diagrams.Automation Level.WebHome"/}}
	275	+{{include reference="FactHarbor.Archive.FactHarbor V0\.9\.23 Lost Data.Specification.Diagrams.Automation Level.WebHome"/}}
250	250
251	251	----
252	252
253	253	== Automation Roadmap Diagram ==
254	254
255		-{{include reference="Test.FactHarborV09.Specification.Diagrams.Automation Roadmap.WebHome"/}}
	281	+{{include reference="FactHarbor.Archive.FactHarbor V0\.9\.23 Lost Data.Specification.Diagrams.Automation Roadmap.WebHome"/}}
256	256
257	257	----
258	258
...	...	@@ -268,4 +268,3 @@
268	268	* [[Requirements (Roles)>>FactHarbor.Specification.Requirements.WebHome]]
269	269	* [[Workflows>>FactHarbor.Specification.Workflows.WebHome]]
270	270	* [[Governance>>FactHarbor.Organisation.Governance]]
271		-