AI Methodology

How Bill Classification Works

Citizen uses two AI models in our classification pipeline: xAI Grok for initial bill classification, and Anthropic (Claude) for an independent audit of xAI's work. This two-model approach is our automated quality check — Claude reviews every xAI classification against the full bill text and flags disagreements. Bills where the two models disagree persistently are labeled as disputed.

Classification Labels

• AI-classified

  Standard AI classification with full bill text analysis.

• AI-classified (partial analysis)

  Bill text exceeded processing limits. Classification based on available text.

• Independently audited

  Classification verified by a second AI model.

• Classification under review

  Two AI models disagreed. Under manual review.

• Historical data

  Pre-AI data not yet classified.

How Governor Alignment Is Scored

Your governor's alignment score is calculated from their pattern of signing and vetoing bills classified by issue. When your governor signs a bill, they are endorsing its policy direction. When they veto one, they are rejecting it. We weight each action by the bill's classification on the issues you care about, then compare it to your preferences — the same math we use for legislators.

What's Not Yet Included

Executive orders — standalone actions your governor takes without the legislature — are not yet factored into this score. We're working on adding them in a future update. For now, your governor's score reflects their legislative record only.

Executive Action Weights

• Signed+1.0
• Allowed without signature+0.5
• Line-item vetoed−0.5
• Vetoed−1.0
• Pocket vetoed−0.5