Adversarial Disconfirmation Pipeline
When you ask an LLM to verify its own output, it confirms what it already said. Single-model pipelines share blind spots. Hallucinations are syntactically indistinguishable from accurate outputs — and confidence scores don't help. The only way to surface fabrication is to challenge claims externally with independent evidence.
The SolutionA 6-stage claim verification pipeline grounded in Popperian falsificationism. Extracts claims, subjects each to adversarial disconfirmation under epistemic isolation using a separate model family, verifies citations against external academic databases, reconciles findings with tiered evidence quality, and generates a self-contained analysis report — stress-testing what survives scrutiny rather than confirming what sounds right.
DECON runs once and reports honestly on what it found. Each stage checkpoints via Lambda Durable Functions before passing to the next.
If the user submits a query instead of pre-existing text, Claude Sonnet 4.6 generates a response for analysis. If text is provided directly, this stage is skipped entirely. Temperature 0.7 for natural output.
Claude Haiku 4.5 reads the full text and extracts every discrete factual claim as a neutrally-framed atomic statement. Each claim is typed (factual, statistical, temporal, causal, comparative, existential, or attributed assertion) and stripped of hedging, framing, and confidence language. Temperature 0.0 for deterministic extraction.
DeepSeek V3.2 (via AWS Bedrock) receives ONLY claim_id, claim_text, and claim_type for each claim. No original text, no generator reasoning, no surrounding context. Claims are processed in batches of 5 to stay within DeepSeek's 8,192-token output limit. Each claim receives a verdict, reasoning chain, counterevidence with quality tier, entity verification status, and recommendation.
Citations produced by the disconfirmer are verified against external academic databases. CrossRef (primary) and Semantic Scholar (fallback) are queried for each unique citation by DOI, author-title match, or fuzzy search. Each citation is classified as confirmed, partial match, or unverified. Results are injected back into the disconfirmation report before reconciliation.
Claude Sonnet 4.6 receives the original text and the citation-enriched disconfirmation report. It weighs counterevidence by quality tier, cross-references citation verification status, identifies cases where the disconfirmer may have erred, performs pattern synthesis across all findings, and produces an executive summary, per-claim results, corrections needed, and overall confidence score.
Claude Sonnet 4.6 transforms the reconciliation output into a self-contained HTML report with executive summary, per-claim analysis with citation verification badges, evidence quality tiers, and APA-formatted references. Reports are stored in S3 and accessible via the client portal.
Claude models are accessed via the Anthropic API directly. DeepSeek is accessed via AWS Bedrock's Converse API. The routing logic detects the model ID and dispatches to the appropriate API.
Evaluation suite across three categories: supported text, fabricated claims, and mixed content containing both real and fabricated claims.
DECON uses Lambda Durable Functions rather than Step Functions. The entire pipeline runs as readable Python code with automatic checkpointing — not ASL JSON state machine definitions.
The Heavy Engine extends the Light pipeline for PDF document analysis. Designed and coded but not yet deployed. When the two independent disconfirmers disagree, the disagreement is flagged as a finding in its own right.
Haiku 4.5 parses uploaded PDF documents into structured text for claim extraction.
Sonnet 4.6 extracts claims with a 0–50 priority scoring matrix across relevance, verifiability, impact, uniqueness, and dependencies.
Primary: Claude Opus 4.6 for Priority 1–2 claims. Secondary: DeepSeek V3.2, independent of the primary. Disagreements flagged prominently.
Sonnet 4.6 reconciles with disagreement tracking. Final stage generates a formatted HTML report for downloadable output.
Discuss how DECON can be integrated into your content workflow — whether you're validating AI-generated reports, auditing research, or building trust into published outputs.