Self-Understanding and Method
The standard is laid open. Measure us by it.
An institute that demands scrutiny must first lay out its own method. Here is how the work is done and what counts as evidence.
Four rules
No title entitles.
Neither ours nor those of the people we cite. A Nobel Prize doesn't make a statement true — it makes it worth checking. Checking happens at the source.
Every figure traces back to a named primary source.
With the date of the last check. Where a rate is stated, the sample and the test setup accompany it. Where a source makes a caveat, the caveat stays with it.
The counter-finding is published alongside it.
What was checked and dismissed stays visibly attached. A collection that only shows confirmations is not a review.
Corrections are dated.
Nothing disappears quietly. Whoever finds an error gets named. The address is open →
What this institute does not do
No predictions presented as facts. Nobody knows which scenario will occur. What is documented is called documented. What is a conclusion is called a conclusion.
No product recommendations. Resilience is not a shopping list. We name the weak point, not a product.
No perspective of the machine. Shutting down a computer doesn't threaten it. Language that casts the system as the victim is not adopted here.
No dramatization beyond the finding. The edge comes from the substance, not from the adjective.
The burden of proof, stated explicitly
Where a development is measurably moving in one direction, the burden of proof falls on whoever claims it will flatten out on its own. "The curve will break at some point" remains an article of faith as long as no one says how you'd recognize the break.
Geneva Institute for ASI Resilience · Geneva · Legal Notice · contact@ASIresilience.org