AKEL Architecture Mermaid

graph TB
    START[User Submits Claim] --> PARSE[Parse & Extract
Components
2-3 seconds]
    PARSE --> PARALLEL{Parallel
Processing}
    PARALLEL -->|Branch 1| EVIDENCE[Evidence Gathering
Web + DB search
5-10 seconds]
    PARALLEL -->|Branch 2| SOURCES[Source Checking
Track record lookup
2-3 seconds]
    PARALLEL -->|Branch 3| SCENARIOS[Extract Scenarios
From evidence context
3-5 seconds]
    EVIDENCE --> VERDICT[Synthesize Verdict
Compile evidence per scenario
2-3 seconds]
    SOURCES --> VERDICT
    SCENARIOS --> VERDICT
    VERDICT --> RISK[Calculate Risk
Assess harm & controversy
1-2 seconds]
    RISK --> PUBLISH[Publish Immediately
No approval gates]
    PUBLISH --> METRICS[Update Metrics
Background]
    LLM[LLM APIs
Claude/GPT/Local
Orchestration Layer] -.->|Used by| PARSE
    LLM -.->|Used by| EVIDENCE
    LLM -.->|Used by| SCENARIOS
    LLM -.->|Used by| VERDICT
    LLM -.->|Used by| RISK
    style PARSE fill:#ffcccc
    style EVIDENCE fill:#ccffcc
    style SOURCES fill:#ccffcc
    style SCENARIOS fill:#ccffcc
    style VERDICT fill:#ffcccc
    style RISK fill:#ffcccc
    style PUBLISH fill:#ccccff
    style LLM fill:#ffffcc

AKEL Pipeline with Parallel Processing - Parse claim (2-3s), then process evidence/sources/scenarios in parallel (max 5-10s), synthesize verdict (2-3s), calculate risk (1-2s). Total: 10-18 seconds (vs 15-30s sequential). Multiple LLM providers supported via orchestration layer for redundancy and cross-checking. No approval gates - immediate publication.