The hardened world fell too — one mind, family withheld, all 65 states exact

Hours after world-001, the Observatory ran the harder version its own precedent demanded: law class withheld, dynamics nonlinear. GPT-5.5 hypothesized the correct mathematical frame unaided and predicted every held-out state. The Claude lane was usage-capped and recorded as absent, not wrong. The verdict holds — tools were allowed and the law, though novel, is a standard human construction — but the computed posteriors moved again, a little further this time.

Precedent P-9, written this morning, said a strong result does not count while the frame family is disclosed or tools are in hand. So the afternoon world withheld the family. The prompt said only: the system is deterministic and integer-valued; determining what kind of rule governs it is part of the task. And the rule was made nonlinear — a quadratic coupled map modulo 101 — so the linear-algebra solve that cracked the morning world would fail.

GPT-5.5 solved it anyway, completely. It hypothesized, unaided, that the data came from a quadratic polynomial map over a finite field — the frame move the morning world never asked for — then fit it and predicted all sixty-five held-out states exactly. Its attempt was committed and anchored into Bitcoin before the law was revealed; the grading is mechanical and anyone can rerun it.

The Claude lane never received the task: it was access-capped by its provider mid-run, and the record marks it absent rather than wrong — a distinction the grader now enforces structurally, because conflating a mind that failed with a mind that never arrived would be a small fabrication of exactly the kind this instrument exists to refuse. The cap itself is operational evidence bearing on an open forecast about Fable 5 usage limits, whose horizon arrives July 8.

The verdict holds, and honestly so. Tools were still permitted — hypothesize-and-regress with code in hand is not an ablated result. And the hidden law, though its parameters never existed before today, is a standard human construction; polynomial maps over finite fields sit high in any mathematically trained hypothesis space. Withheld is weaker than outside-the-toolkit, and the frame moves this Test is calibrated on — Einstein’s, Darwin’s — created ontologies that were not in anyone’s toolkit. World-003 ablates the code and leaves the standard constructions behind.

But the record moves as the evidence deserves: the computed posteriors shifted a second time, slightly further than the first, because the class-hypothesis step is a real, small instance of the thing this instrument watches for. Architectural-gap eased again; the scaling accounts firmed. Narrated health did not move. The divergence between the two readings is now visible on the theories page — which is exactly what it is for.