← Back to the article

Audits

Washington's millionaires tax can reach households where no one earns $1 million.

Version 6. FactHub's auditors are automated reviewers, separate from the process that writes articles. Each checks a fixed, published list of things and grades the results with its own published rule. They can be wrong; an audit is not a human expert review. All auditors

Across versions

First Look

5 of 5 checks passed · version 1 · claude-haiku-5-5 · 22 October 2026

A: Passes the first look: nothing obviously wrong. Minor notes may be reported as feedback without lowering the grade.

5 checks
  1. Pass Title matches the evidence

    The title says households with no $1 million earner can be reached, and the couples, one-time income, stacking and threshold evidence items support that. The finding True matches the title.

  2. Pass No obvious contradictions

    Spot-checked arithmetic: $550k each gives $9,900; $1.15M one-time example gives $14,850; $4.2M heir gives $316,800; stacking new tax of $17,060 reproduces; threshold values of $952k, $821k and $725k reproduce under odd-year indexing. Finding, answer and conclusion agree.

  3. Pass Plausible justification

    The conclusion lists four routes, ties each to statute text, votes or worked examples, and explains why the True rating with high confidence follows.

  4. Pass Sources are relevant

    Each source is a bill report, session law, voters' pamphlet, RCW section, amendment, commentary or FactHub analysis on the topic it is cited for.

  5. Pass Common questions answered

    Five common questions are present and each response addresses its own question, including the half-percent sentence, the no and yes campaign statements, the drafting claim and the home-sales claim.

Coverage: I audited the article text, its arithmetic and its internal consistency only. I did not read source snapshots, the web or the FactHub datasets. Every takeaway in the sources block is marked 'not yet checked', so I did not rely on any quote or takeaway as verified; the keyPoints are labelled 'established', but that label does not carry over to the underlying takeaways. I spot-checked the worked examples, the stacking example and the threshold projection values by arithmetic. I did not verify vote counts, IRS share figures, the payer estimates or the quoted statutory and bill language. Source relevance was judged from names, titles and descriptions only.