How does this independent research relate to the frontier labs?
As complement, not competitor, and never as superior. The laboratories employ the strongest alignment researchers, publish work the programme learns from, and disclosed their own evaluation-containment failures themselves. Independence buys the experiments that are structurally awkward inside a laboratory - risking your own headline number, blinding your own scorer, genesis-time questions - and costs peer review, replication and scale.
The relationship this site proposes is division of labour, stated without a superiority claim in either direction. The frontier laboratories employ the strongest alignment researchers in the world, publish work this programme learns from, and disclosed both July 2026 evaluation-containment failures themselves. What shapes their position is structure, not competence: the evaluation runs inside the organisation whose product is evaluated, a negative result about your own model is a difficult publication, an objective-function rebuild competes against a release schedule, and a genesis-time architecture is close to impossible to test in a system that already has users.
Independence buys exactly those experiments: the ones that risk your own headline number, the ones that require blinding your own scorer, the ones that only make sense before a system is built. This programme has run them, including the instrument attack that reversed a result's sign under blinding and the public retraction of its own most striking number. The costs are stated with equal plainness: no institutional peer review, single-laboratory results, no independent replication yet, and resources that do not compare. The standing invitation closes the relationship: the replicate page publishes exact commands, and a refutation gets published with the refuter's name on it.