Methodology

From raw source material to a traceable public report.

HSI runs a five-stage pipeline. At every stage, one rule holds: every claim must be traceable back to its source.

  1. Step 1

    Raw source material

    Public posts, articles, transcripts, and campaigns are captured and archived with timestamps and cryptographic hashes.

  2. Step 2

    AI extraction

    Models extract claims, entities, and rhetorical structures from raw material — always preserving a link to the exact source passage.

  3. Step 3

    AI interpretation

    A structured classifier interprets framing, technique, and influence patterns, attaching a confidence score to every inference.

  4. Step 4

    Human review

    Trained reviewers verify, challenge, or reject each interpretation. Models assist; humans make the published call.

  5. Step 5

    Public report

    Only verified, fully-traceable findings are published — with the evidence chain available for public inspection.

Every claim must be traceable.

This is the non-negotiable foundation of HSI. If a conclusion cannot be walked backward to a verifiable source, it is not published.

Confidence, not certainty

Every AI inference carries an explicit confidence score so reviewers and readers can weigh it appropriately.

Documented limitations

We publish what our models can and cannot do, including known failure modes and category boundaries.

Reproducible analysis

Archived evidence and versioned models mean findings can be re-examined and independently audited.