CONCEPT
Designing for Surprise
The structural prescription Campbell's framework implies — systems, workflows, and institutions that generate blind variation as a byproduct of their operation rather than requiring it as a deliberate sacrifice of exploitation efficiency.
Designing for surprise means creating the conditions under which blind variation occurs within and alongside the exploitative workflow that AI enables. The design principle derives from Campbell's framework: genuine exploration requires blind variation; blind variation requires conditions where directed variation's constraints are relaxed; those conditions must be produced structurally because individual intention cannot sustain them against organizational pressure. The beaver does not choose which eddies the dam creates — the beaver builds the dam, and the dam creates the eddies. The deliberation is in the placement of resistance; the blindness is in what the resistance produces. Applied to AI-augmented work, designing for surprise means engineering tools, workflows, organizations, and institutions that deliberately introduce perturbation — structured encounter with the unexpected — into the exploitative flow.
In The You On AI Field Guide
The most straightforward design implication concerns the AI tools themselves. Current language models are optimized for plausibility — outputs that conform to training data regularities. A more epistemologically