CONCEPT
Task Seepage
The Berkeley researchers' term for the colonization of previously protected temporal spaces by AI-accelerated work — the mechanism through which the recovery windows of pre-AI workflows disappear.
Task seepage describes the pattern documented in the 2026 Berkeley ethnographic study of AI-augmented organizations: workers prompting during lunch breaks, refining outputs during elevator rides, testing features in the minutes between meetings, filling every available temporal gap with productive AI interaction. The researchers named the phenomenon to describe behavioral colonization of rest periods. Selye's framework identifies what the behavior costs: the elimination of the micro-recovery windows that the stress response requires to remain adaptive. Each gap filled with productive interaction is a gap during which cortisol does not decline, the parasympathetic system does not activate, and the default mode network does not perform the consolidation that converts processing into learning. Task seepage is, in biological terms, the conversion of a cyclical stress pattern into a linear one — and cyclical structure is what keeps demand from becoming damage.
In The You On AI Field Guide
The Berkeley study by Xingqi Maggie Ye and Aruna Ranganathan, published in Harvard Business Review in February 2026, documented the pattern through eight