FactHub auditor: Overall Review
Findings
Everything this auditor has flagged, across all its versions, and what happened next: still open, applied in a later version, acknowledged, or rejected as a false alarm. A finding raised again by a later run is superseded by the newer one, so each problem is counted once.
74 findings: 61 open, 13 applied.
-
Version 4 reports the all-funds operating budget from the dataset's ngf-by-biennium table: $68.8 billion in 2013-15 to $142.4 billion in 2023-25, up 106.9%. The figure is in the budget-doubled evidence, key points, context and conclusion. The context and method now explain why the Near General Fund is the main measure.
-
In version 3, the conclusion, rents evidence and rents objection label the 4th-most unsheltered rise as counted. They add that it stays 4th with the central method adjustment and falls to 21st at the high adjustment. The conclusion now says whether King County did worse than most peers depends on the method effect: it did under the central estimate, but not the high one.
-
Version 3 states the 2015 count-area expansion and the 2017 Count Us In change in the context, count evidence, method objection and conclusion, with the count within each method span. The 2015-vs-2005 comparison in the promises evidence now notes that the 2015 count covered a larger area, so the comparison is approximate. The answer no longer says "roughly doubled".
-
Title and answer softened in version 4: IQ studies concern total intake, not toothpaste. Key points note that 1.5 mg/L is about twice the U.S. level and that the evidence below it is unsettled.
-
Version 4 notes that the FDA cited the IQ meta-analysis and gut-microbiome concerns. It also adds the neurotoxin objection, the Levy and Lanphear editorials, and the label rule that 1,500 ppm paste is not for under-6s.
-
Version 4 adds the Texas Attorney General's September 15, 2025 release (archived) as the primary source for the Colgate agreement. DentistryIQ is kept as a secondary account.
-
The article now explains that "no unreasonable risk" is a regulatory judgement, not zero risk, in a key point, the EPA evidence item and the conclusion. It also notes that EPA's published determination does not combine laundry exposure with other sources, and it adds P&G's North America compliance statement.
-
The settlement now cites the California Attorney General's record of the case (San Francisco Superior Court No. CGC-12-517674: judgment January 22, 2013, 25 ppm limit). Newsday is credited as the publisher of the testing report, and method step 4 now says so.
-
The daily-dose statement was removed: the California Attorney General's record shows the settlement set a 25 ppm product limit, which is now cited. The unsourced claim about 'natural' products was removed; the response now relies only on the cited ethoxylation source and New York's limit.
-
The court challenge is now sourced (Union Carbide Corp. v. EPA, No. 24-60615, 5th Cir., via Beveridge & Diamond, January 2026), with EPA's 2025 reconsideration and the stay.
-
The article now explains that "no unreasonable risk" is a regulatory judgement, not zero added risk. It also says that EPA's published determination does not describe combining laundry exposure with other products and tap water, citing EDF's critique.
-
Now cites the California Attorney General's Proposition 65 record of the case (No. CGC-12-517674), which gives the 25 ppm limit and the September 30, 2013 deadline. As You Sow is kept only for the nationwide scope.
-
The unsourced claim about 'natural' products was removed. The ethoxylation background is supported by the cited EPA and Holland & Knight sources, and the response now relies only on that and New York's limit.