# Knowledge Decay Knowledge decay is the process by which stored knowledge loses accuracy, relevance, or usefulness over time. It applies to human memory, organizational knowledge bases, documentation, and AI context systems alike. The [[Law of staleness]] describes the mechanism: information's value declines rapidly as it ages. Technical documentation becomes outdated as APIs change. Best practices evolve. Facts get superseded. Notes about a project's architecture describe last year's system, not today's. In [[Personal Knowledge Management (PKM)]], knowledge decay is why periodic review matters. A note written two years ago may contain outdated assumptions. A [[Zettelkasten method|Zettelkasten]] that's only written to but never revisited accumulates decay silently. The same applies to organizational [[Knowledge Management (KM)]]: institutional knowledge decays as people leave, processes change, and context gets lost. Knowledge decay is the broader concept that specific forms of rot extend from: - [[Bit rot]]: digital data corruption from storage degradation - [[Link rot]]: URLs breaking as pages move or disappear - [[AI Context Rot]]: AI instructions and memory becoming stale relative to the system they describe The AI angle is particularly interesting: AI context is just another knowledge system, subject to the same decay dynamics as any other. [[Context Hygiene]] and [[Context Drift]] awareness are the AI-specific mitigations, but they're really just applications of the same PKM principles: review, update, prune, connect. ## References - ## Related - [[Law of staleness]] - [[Bit rot]] - [[Link rot]] - [[AI Context Rot]] - [[Context Drift]] - [[Context Hygiene]] - [[Knowledge Management (KM)]] - [[Personal Knowledge Management (PKM)]] - [[Enterprise Knowledge Management (EKM)]]