Evaluator Record
One record per evaluator: what each was engaged to test, on what terms, what it found in the document's own words, and what happened afterwards, with the money and seats behind it. Every line is a store record and every record links to the document it was read from.
A lab’s safety claim names an outside tester, and the tester’s report names its terms, or does not. These pages keep the ledger per evaluator: every engagement a lab document records, the access window, model stage, publication right and payment that document states, the finding quoted, and the events on the record afterwards, from an incident on a model it tested to a lab’s statement that it acted or did not act on what it found. A term the document is silent on reads not stated, and each table counts how many of its rows state each field. The records are drawn from the same store as the Evaluator Observatory; the narrative is on The Evaluator Alliances.
Filters combine: any of the values ticked within a dimension, all of the dimensions at once. A table that has no column for a dimension you have set says so instead of showing rows that cannot match. The address bar carries the slice.
The records 38 evaluators, store as of 2026-10-18
Ordered by engagements on record, then outcomes, then money rows. Open a record for the timeline, the terms grid, the findings quoted and what followed; a lab name narrows the list to the evaluators its documents name.
Dated 125 of 125Model named 117 of 125Terms stated 66 of 125Finding quoted 53 of 125
Access window 21 of 125Model stage 26 of 125Publication 11 of 125Payment 6 of 125Finding, quoted 53 of 125
Outcomes: Dated 18 of 18Outcomes: Model named 12 of 18Outcomes: Tied to an engagement 13 of 18Outcomes: Evaluator named by the record 18 of 18