Place the evidence you were sent
What rung is it on?
You are holding something a vendor published: a benchmark page, a trust center, a report, a claim. Answer from what is actually in front of you. If you cannot tell, say so, and the ladder will not move up on your behalf. Nothing you answer leaves this page.
step 2 · rung 2 test
Is a named, publicly documented benchmark, standard, or evaluation identified?
A name a third party could look up on their own, for example TOFU, MUSE, WMDP, AILuminate, SOC 2, ISO/IEC 42001. An internal, unnamed evaluation does not count.
Proof Registry holds 34 published artifacts on this same ladder, including ours and including where a competitor beats us. Start over.