CONCEPT
Plausibility Corruption
The specific epistemic hazard AI introduces — output optimized to sound right rather than to be right, producing confident simulation of expertise that passes surface evaluation while lacking the foundation to catch its own errors.
Plausibility corruption names the specific mechanism through which AI-mediated output undermines the evaluation systems on which knowledge work depends. AI systems are trained, by their fundamental structure, to produce output that matches the distribution of competent human output — output that sounds right, reads well, matches the patterns human evaluators recognize as expertise. This surface optimization is not a bug but the tool's core capability. The hazard is that plausibility is a surface property, and evaluation conducted primarily at the surface cannot distinguish between output grounded in understanding and output matching the patterns of understanding. The confident simulation passes. The simulation's absence of foundation is revealed only when reality administers a test the plausibility did not anticipate.
In The You On AI Field Guide
Crawford's framework reveals plausibility corruption as structurally distinct from older epistemic hazards. Deliberate deception requires an agent with intent to deceive. Bullshit, in Harry Frankfurt's technical sense, requires indifference to truth. Plausibility corruption requires neither.