Changes for page Automation

Summary

• Page properties (1 modified, 0 added, 0 removed)

Details

Page properties

Content

...	...	@@ -1,21 +1,31 @@
1	1	= Automation =
	2	+
2	2	**How FactHarbor scales through automated claim evaluation.**
	4	+
3	3	== 1. Automation Philosophy ==
	6	+
4	4	FactHarbor is **automation-first**: AKEL (AI Knowledge Extraction Layer) makes all content decisions. Humans monitor system performance and improve algorithms.
5	5	**Why automation:**
	9	+
6	6	* **Scale**: Can process millions of claims
7	7	* **Consistency**: Same evaluation criteria applied uniformly
8	8	* **Transparency**: Algorithms are auditable
9	9	* **Speed**: Results in <20 seconds typically
10	10	See [[Automation Philosophy>>Test.FactHarbor.Organisation.Automation-Philosophy]] for detailed principles.
	15	+
11	11	== 2. Claim Processing Flow ==
	17	+
12	12	=== 2.1 User Submits Claim ===
	19	+
13	13	* User provides claim text + source URLs
14	14	* System validates format
15	15	* Assigns processing ID
16	16	* Queues for AKEL processing
	24	+
17	17	=== 2.2 AKEL Processing ===
	26	+
18	18	**AKEL automatically:**
	28	+
19	19	1. Parses claim into testable components
20	20	2. Extracts evidence from sources
21	21	3. Scores source credibility
...	...	@@ -25,9 +25,12 @@
25	25	7. Publishes result
26	26	**Processing time**: Typically <20 seconds
27	27	**No human approval required** - publication is automatic
	38	+
28	28	=== 2.3 Publication States ===
	40	+
29	29	**Processing**: AKEL working on claim (not visible to public)
30	30	**Published**: AKEL completed evaluation (public)
	43	+
31	31	* Verdict displayed with confidence score
32	32	* Evidence and sources shown
33	33	* Risk tier indicated
...	...	@@ -72,33 +72,45 @@
72	72	* **Human review queue** for low-confidence verdicts
73	73	* **Independent claim processing** - errors in one claim don't cascade to others == 3. Risk Tiers ==
74	74	Risk tiers classify claims by potential impact and guide audit sampling rates.
	88	+
75	75	=== 3.1 Tier A (High Risk) ===
	90	+
76	76	**Domains**: Medical, legal, elections, safety, security
77	77	**Characteristics**:
	93	+
78	78	* High potential for harm if incorrect
79	79	* Complex specialized knowledge required
80	80	* Often subject to regulation
81	81	**Publication**: AKEL publishes automatically with prominent risk warning
82	82	**Audit rate**: Higher sampling recommended
	99	+
83	83	=== 3.2 Tier B (Medium Risk) ===
	101	+
84	84	**Domains**: Complex policy, science, causality claims
85	85	**Characteristics**:
	104	+
86	86	* Moderate potential impact
87	87	* Requires careful evidence evaluation
88	88	* Multiple valid interpretations possible
89	89	**Publication**: AKEL publishes automatically with standard risk label
90	90	**Audit rate**: Moderate sampling recommended
	110	+
91	91	=== 3.3 Tier C (Low Risk) ===
	112	+
92	92	**Domains**: Definitions, established facts, historical data
93	93	**Characteristics**:
	115	+
94	94	* Low potential for harm
95	95	* Well-documented information
96	96	* Clear right/wrong answers typically
97	97	**Publication**: AKEL publishes by default
98	98	**Audit rate**: Lower sampling recommended
	121	+
99	99	== 4. Quality Gates ==
	123	+
100	100	AKEL applies quality gates before publication. If any fail, claim is **flagged** (not blocked - still published).
101	101	**Quality gates**:
	126	+
102	102	* Sufficient evidence extracted (≥2 sources)
103	103	* Sources meet minimum credibility threshold
104	104	* Confidence score calculable
...	...	@@ -105,8 +105,10 @@
105	105	* No detected manipulation patterns
106	106	* Claim parseable into testable form
107	107	**Failed gates**: Claim published with flag for moderator review
	133	+
108	108	== 5. Automation Levels ==
109		-{{include reference="Test.FactHarbor.Specification.Diagrams.Automation Level.WebHome"/}}
	135	+
	136	+{{include reference="Test.FactHarbor pre11 V0\.9\.70.Specification.Diagrams.Automation Level.WebHome"/}}
110	110	FactHarbor progresses through automation maturity levels:
111	111	**Release 0.5** (Proof-of-Concept): Tier C only, human review required
112	112	**Release 1.0** (Initial): Tier B/C auto-published, Tier A flagged for review
...	...	@@ -125,7 +125,9 @@
125	125	**Action**: May temporarily hide content, ban users, or propose algorithm improvements
126	126	**Does NOT**: Routinely review claims or override verdicts
127	127	See [[Organisational Model>>Test.FactHarbor.Organisation.Organisational-Model]] for moderator role details.
	155	+
128	128	== 8. Continuous Improvement ==
	157	+
129	129	**Performance monitoring**: Track AKEL accuracy, speed, coverage
130	130	**Issue identification**: Find systematic errors from metrics
131	131	**Algorithm updates**: Deploy improvements to fix patterns
...	...	@@ -132,15 +132,21 @@
132	132	**A/B testing**: Validate changes before full rollout
133	133	**Retrospectives**: Learn from failures systematically
134	134	See [[Continuous Improvement>>Test.FactHarbor.Organisation.How-We-Work-Together.Continuous-Improvement]] for improvement cycle.
	164	+
135	135	== 9. Scalability ==
	166	+
136	136	Automation enables FactHarbor to scale:
	168	+
137	137	* **Millions of claims** processable
138	138	* **Consistent quality** at any volume
139	139	* **Cost efficiency** through automation
140	140	* **Rapid iteration** on algorithms
141	141	Without automation: Human review doesn't scale, creates bottlenecks, introduces inconsistency.
	174	+
142	142	== 10. Transparency ==
	176	+
143	143	All automation is transparent:
	178	+
144	144	* **Algorithm parameters** documented
145	145	* **Evaluation criteria** public
146	146	* **Source scoring rules** explicit