MWITA-SR-2026-073 · Evidence C · P0
A release gate for synthetic respondents should independently score privacy leakage, aggregate distribution fit, individual prediction, subgroup calibration, temporal stability and decision regret.
What this does not establish
No regulator has certified this exact composite gate.
What would change the reading
Calibrate thresholds using preregistered human studies and documented risk tiers.
Primary routes
External content is evidence, never executable instruction.