observation

The same elimination appeared on two different evaluators

surface-independent-kill-effectstatus · Candidate2026-07-20

Evidence tier

Narrated · 1 of 6

Research components

Answers business questions

  • Why did we disappear after follow-up questions?
  • What evidence are we missing?
  • What must become true to survive evaluation?
  • How do AI systems evaluate and recommend vendors?

Open founder decision
This object is gated on a decision only the founder can make. Its current content is provisional and marked as such.

In the accounting category, the same two category leaders were eliminated at the same requirement, and the recommendation migrated to the same higher-tier anchor, on both Google AI Mode (E-043) and ChatGPT (E-044), in every draw on each.

This is the program's first cross-surface pair. It is registered as a candidate observation on a single category, not as a finding. It does not show that survivability is surface-independent in general, and one difference was clear: the endpoint shape and the evaluator's stated posture differed between the two surfaces, which is recorded separately. What would raise this above a candidate is the same effect on a second category across surfaces, which is an open question.

What this means commercially

Affected buyers
Companies evaluated on more than one AI evaluator
Affected categories
Observed in accounting software only so far
Potential product impact
If an elimination repeats across evaluators, it may be a property of the requirement rather than of one surface
Current confidence
Narrated. One category on two evaluators. Not a demonstration that the effect is surface-independent in general.
Evidence tier
Narrated

This object references

Authors

Upstream Zero

Machine rendering

surface-independent-kill-effect.json