every ai answer about you, cross-examined against its own sources

generative search now answers questions about companies directly and cites sources, and nobody checks whether those sources actually support what it says.

recorded 2026-09-01 · 148 searches · 8 markets

01

the case that started this

In 2025 a Google AI Overview told searchers a Minnesota solar installer was being sued by the state Attorney General. It was not. The answer cited four sources and none of them contained the claim. The company reported losing a $150,000 contract and sued Google; that litigation is ongoing.

Google has since corrected that answer, which is the point — generative answers change underneath you, so this has to be measured continuously rather than once.

cited
four sources, formatted exactly like a grounded answer
contained
none of them carried the claim
cost
a $150,000 contract
02

what nobody measures

There is a whole category of AI-visibility tools and they all measure the same thing — is the brand mentioned, in which prompts, with what sentiment.

None of them check whether the sources cited beside a sentence actually contain that sentence. The missing property is attribution integrity.

03

how a deposition runs

  1. 01docketing

    Target entity is registered, aliases and collision-set neighbours are indexed, and the search budget is established.

  2. 02line of questioning

    Deterministic probe families are compiled across identity, adverse, commercial, qualification, and operational queries.

  3. 03deposition

    Probes are dispatched across a geographic locale grid, capturing append-only raw engine observations and suppression records.

  4. 04claim decomposition

    Generative answers are parsed into atomic factual assertions mapped to their containing block text and cited reference indices.

  5. 05cross-examination

    Every atomic assertion is tested against the union of its block citations via SerpApi snippet evaluation and verified page retrieval.

  6. 06divergence and scoring

    Observed assertions are clustered across locales and engines to compute attribution integrity, consistency, and overall grounds scores.

04

the eight verdicts

grounded
Cited sources support the claim and independent corroboration checks out.
miscited
The claim may be true, but the sources Google cited do not contain it.
unsourced
Cited sources are silent on the claim and no independent corroboration exists.
contradicted
Independent corroborating sources directly refute the assertion.
stale
The assertion was previously true, but a newer authoritative source refutes it.
conflated
The assertion is true of a different entity in the collision set.
unverifiable
Cited sources could not be read due to paywalls, blocks, or unparseable formats.
opinion
Subjective value judgement or prediction, excluded from scoring.

UNVERIFIABLE is never counted as a defect, because an unreadable source must never become an accusation.