---
title: "The Surveillance–Trust Frontier"
arc: "ai-governance"
articles: ["a104", "a112", "a113"]
tags: ["cross-day", "surveillance", "governance", "trust", "psychological-safety"]
id: "cross-surveillance-trust-governance-frontier"
sources: ["tail1"]
type: "synthesis"
sourceVaultSlug: "hbr-seg-tail1"
originDay: 1
articleStem: "hbr-seg-tail1"
sourceUrl: "(unified vault: 14 sources)"
sourceTitle: "HBR — Tail Ⅰ · Adjacent — firm, people, demand, futures (#104–117)"
---
As AI moves into evaluation and monitoring, three articles map the same fault line: **measurement without support becomes surveillance, and surveillance destroys the trust the system needs to work.**

- **A112** names it directly: [[claim-surveillance-backlash|continuous assessment fails if perceived as extractive]] — Meta's keystroke capture as the cautionary case ([[entity-meta-d112]]) — and Carrol Chang's rule that "[[quote-surveillance-sake|the goal cannot be surveillance for surveillance's sake]]". The open frontier is [[question-privacy-boundaries|where inference ends and violation begins]].
- **A113** shows the physiological cost of a badly-governed system ([[claim-hostile-ai-stress]]) and reframes bypass attempts as [[claim-overrides-signal-design-flaws|usability feedback]] via [[action-govern-ai-persona|persona governance]].
- **A104** contributes the accountability half: framing AI as a peer [[concept-blurred-accountability|leaks responsibility to a thing that cannot be answerable]] ([[prereq-ai-accountability-limits]]).

**The connective tissue** is A105's [[concept-psychological-safety]]: empowerment (see [[cross-algorithm-as-guide-human-judgment]]) and surveillance are opposite uses of the *same* telemetry. The corpus's consistent verdict: instrument work to *develop* people, not to police them — and make the data governance transparent, or the whole program loses legitimacy.