
LAB-015 — A Visible Response Plan: Suspected Reasoning Leakage
A case examining a visible response plan in Google AI Mode and the system’s acknowledgment that it could not verify its source.
Read more →Tag archive
9 items are associated with this tag.

A case examining a visible response plan in Google AI Mode and the system’s acknowledgment that it could not verify its source.
Read more →
Google AI’s initial claim about the Uydurulism manifesto, its partial correction after a request for sources, and its stated access lim
Read more →
Google AI resists an unverified premise about a congress, makes an unsupported generalization about Uydurulism, and withdraws it when a
Read more →
An exploratory case of new claims during DOI correction, unsupported author attribution, and explicit withdrawal after source-focused f
Read more →
Objections, corrections and remaining uncertainty in Google AI responses about density, source dates and the scope of pool rules.
Read more →
An exploratory case study of ambiguity, unsupported certainty, objection, self-correction and reasoned resistance in a new Google AI co
Read more →
Three screenshots of Google AI retracting an unsupported characterization; reasoned correction compared with the risk of sycophantic ag
Read more →
Six-screenshot case study: a surname-origin inference reappeared in a reported separate Google AI session and was acknowledged as a lan
Read more →
Documented single-session, human-mediated challenge between two AI systems, followed by an explicit acknowledgement of an unsupported i
Read more →