{"id":"surface-independent-kill-effect","type":"observation","title":"The same elimination appeared on two different evaluators","created":"2026-07-20","status":"Candidate","authors":["Upstream Zero"],"edges":[{"rel":"part-of","to":"vendor-elimination"},{"rel":"part-of","to":"validation-and-evidence"},{"rel":"derives-from","to":"E-043"},{"rel":"derives-from","to":"E-044"}],"tier":"Narrated","commercialRelevance":{"affectedBuyers":"Companies evaluated on more than one AI evaluator","affectedCategories":"Observed in accounting software only so far","potentialProductImpact":"If an elimination repeats across evaluators, it may be a property of the requirement rather than of one surface","currentConfidence":"Narrated. One category on two evaluators. Not a demonstration that the effect is surface-independent in general."},"observationType":"behavioral","founderDecision":"Whether this counts as first cross-surface evidence for the pattern, or how it should be weighed toward promoting the battery, is a pending founder decision. It is registered here as a candidate on one category, not as an established finding.","pubState":"published","body":"In the accounting category, the same two category leaders were eliminated at\nthe same requirement, and the recommendation migrated to the same higher-tier\nanchor, on both Google AI Mode ([E-043](/experiments/E-043)) and ChatGPT\n([E-044](/experiments/E-044)), in every draw on each.\n\nThis is the program's first cross-surface pair. It is registered as a candidate\nobservation on a single category, not as a finding. It does not show that\nsurvivability is surface-independent in general, and one difference was clear:\nthe endpoint shape and the evaluator's stated posture differed between the two\nsurfaces, which is recorded separately. What would raise this above a candidate\nis the same effect on a second category across surfaces, which is an open\nquestion.","url":"/observations/surface-independent-kill-effect","machineUrl":"/objects/surface-independent-kill-effect","referencedBy":[],"_meta":{"site":"Upstream Zero · Commercial Intelligence for AI-Mediated Commercial Evaluation","version":"0.1","note":"Claims are presented at their evidence tier; Narrated is the lowest. Verify by walking edges, not by trusting us."}}