Openness makes consequential claims testable. Without evidence, we cannot distinguish systemic risk from ordinary engineering failures used to justify sweeping regulation.

Without evidence, AI safetyism looks like a self-appointed group claiming privileged knowledge to constrain