# Agentrification (Scott Logic) > [!question] Lingering Questions > - Where is the actual line between augmenting (copilot) and displacing (custodian) a human — is "Mixed-Use" a stable state or just a transition? > - Who captures the value that agent efficiency frees up — workers, firms, or the infrastructure vendors collecting the compute "rent"? > - What governance would let a community or codebase deliberately preserve human "friction" and character? > [!tip] Article Ideas > - "The custodian trap" — a short post naming the specific failure mode where agent tooling adds more supervision/prompt-wrangling toil than it removes, aimed at maintainers evaluating AI-assist tools. > - "Utility test vs. status test" — apply the piece's framing to one AI/MCP tool The WPOCC or a WordPress team recently adopted. > [!link] Cross Reference > - [[3. Resources/Permanent Notes/Automation Relocates Toil Into Supervision]] — the vault's existing Permanent note built from this same source; this Literature note supplies its remaining atoms and context. > - [[Open Source Governance MOC]] — maintainer/contributor "legacy resident" framing applies directly to AI-generated issues/PRs norms. > [!quote] Vocabulary > - *agentrification* — the article's coinage: displacement of human character/contribution by AI agents, modeled on urban gentrification (rising "rent," sanitized character, "legacy" stigma). > - *custodian trap* — a worker demoted from producer to supervisor of the agent that replaced them, paying "rent" in compute, latency, and prompt-wrangling. ## Bib capsule Satirical mock-white-paper from a UK software consultancy's engineering blog (2026-06-29), presented as an "Institute for Synthetic Urbanism & Computational Decay" report. Author unlisted (house blog voice). No prior published edition; single web post. ## Author's thesis (what it's about / what's said in detail) The piece uses a sustained gentrification metaphor to argue that autonomous AI agents displace humans from communities, jobs, and codebases much as gentrification displaces residents: pricing out authentic contribution, demoting producers to custodians of their replacements, and evicting lean, working solutions in favor of bloated "agent-native" ones adopted for status rather than utility. A mock "Agentrification Index" scores a system's saturation, friction, and fidelity into tiers from "Legacy" to "Agentrified." The scoring apparatus is openly satirical — the serious payload is the closing line: under full agent optimization, messy, inefficient human contribution may be the most valuable thing left. ## Key arguments (≤3) 1. **Professional displacement is the sharpest, most testable claim**: automation that removes production work often relocates the effort into supervision — "constant debugging, prompt-wrangling, and resource-monitoring" — and that supervision overhead can exceed what was saved. 2. **Algorithmic displacement targets adoption motive, not capability**: lean, working code gets replaced by heavier "agent-native" frameworks chosen for architectural status/hype rather than measured utility. 3. **The whole apparatus is rhetorical scaffolding, not measurement**: the Agentrification Index gives the critique a memorable, falsifiable-sounding shape, but the piece offers illustration, not evidence — the real argument rests on the analogy holding up under scrutiny, which (per the Logical Validity read) it mostly does as a diagnostic lens. ## Claim atoms ### Automation can demote workers to custodians of their replacements > [!quote] ID:rw_agentr_01 > The humans remain, but they no longer build or create; they are now the "superintendents" of an agent-occupied building. Where/why: professional-displacement definition, body paragraph. Names the "custodian trap" — the testable core of the critique, and the basis of the vault's existing Permanent note [[3. Resources/Permanent Notes/Automation Relocates Toil Into Supervision]]. ### Inefficient human contribution becomes the scarce asset under full automation > [!quote] ID:rw_agentr_02 > the most valuable asset remaining may very well be the one that is messy, irrational, and inefficiently human. Where/why: the closing line, after the satire drops its mask — the serious payload of the whole piece. ### Self-automating personas signal a space tipping toward displacement > [!quote] ID:rw_agentr_03 > Keep an eye out for the Pioneer Agentrifier – the user who automates their entire persona. Where/why: mock "auditor instructions" section — a concrete, observable leading indicator for community moderation. ### Agent frameworks add compute "rent" for tasks that need no generation > [!quote] ID:rw_agentr_04 > a 4GB Docker container tha imports ten different agentic orchestration libraries just to print "Hello, World". Where/why: algorithmic-definition "Dependency Bloat" sign (typo preserved verbatim from source). Hyperbolic, but points at real cost/latency/supply-chain risk from status-driven tool adoption. ### Tooling chosen for architectural status rather than utility is a cost, not an upgrade > [!quote] ID:rw_agentr_06 > a desire for architectural status rather than utility Where/why: the algorithmic-displacement definition's causal claim — lean code gets "evicted" because the heavier framework signals something to peers, not because it performs better. Distinct atom from H4 (which is the illustration); this is the underlying mechanism claim, and the basis of the new Permanent note below. ## Permanent candidates - **[[3. Resources/Permanent Notes/Automation Relocates Toil Into Supervision]]** — already exists in the vault (created 2026-08-01, same origin doc). Covers claim atom H1 in full, including the "ironies of automation" (Bainbridge) connection. Not recreated here. - **Tooling chosen for architectural status rather than utility is a cost, not an upgrade** — new candidate from H6/H4, no existing Permanent covers the status-vs-utility adoption mechanism specifically (checked: no match on "architectural status," "résumé-driven," "boring technology"). Promoted below. ## Processing checklist - [x] Checked for an existing Permanent note covering the custodian-trap claim — found one; linked instead of duplicating. - [x] Checked for an existing Permanent note on status-driven tooling adoption — none found; created. - [ ] Revisit if a real (non-satirical) study of agent-adoption supervision overhead surfaces — would let the custodian-trap note graduate past `budding`. - [x] Connected into [[3. Resources/MOCs/AI & the Future of Knowledge Work MOC]] and [[3. Resources/MOCs/Open Source Governance MOC]]. ## Links - [[3. Resources/Readwise/Articles/Agentrification and the Agentrification Index]] — full saved source document - [[3. Resources/MOCs/AI & the Future of Knowledge Work MOC]] — frames agent-driven displacement as a knowledge-work pattern alongside the vault's other automation-and-toil notes - [[3. Resources/MOCs/Open Source Governance MOC]] — the "legacy resident"/maintainer framing is a direct governance concern for AI-generated contributions