causal
Efficacy Tests Mislead Because Best Inputs Aren't Scalable
A/B tests and efficacy tests use the best inputs a researcher can marshal, but those best inputs cannot be replicated at scale; failing to disclose that a result was an efficacy test causes voltage drops when policymakers later try to scale it.
causalpending
Speaker
John ListEvidence Quote
“when you scale, you're not going to be able to employ the best possible inputs.”
Created: 6/13/2026, 12:10:15 AM
Your notes
Loading notes...