Distinguishing the ethical friction deficit from related concepts
| Concept | What it primarily concerns | What the EFD adds |
|---|---|---|
| Organizational silence (Morrison and Milliken, 2000) | Why actors withhold voice – including through structural and climate barriers, not merely individual choice | A structural, designed interruption of the decision chain, independent of whether any individual is willing to speak |
| Voice/psychological safety (Van Dyne and LePine, 1998; Edmondson, 1999) | Whether members can and will raise concerns | Voice may be exercised without power to alter an output already shifted into an executable register; the deficit persists even under high psychological safety |
| Organizational visibility/transparency | What information becomes visible, legible or safely reportable within the organization | A visible decision can still be uncontestable; visibility is not contestability |
| Automation bias (Skitka et al., 2000) | Why individuals over-trust system outputs | The deficit can exist even where no one is biased, because no structural contestation point was designed |
| Human oversight (Siebert et al., 2023) | Whether a human is involved and in what role – including authority commensurate with control | Presence is not capacity: a human may be involved yet hold no authority to suspend or revise before execution |
| Contestable AI (Alfrink et al., 2023) | Whether a decision can be questioned, across the system lifecycle | The organizational loss of effective contestability specifically before execution, when challenge can still prevent embedding |
| Algorithmic accountability (Busuioc, 2021) | Who must answer and how, before a forum – including ex ante duties | Whether anyone could still intervene before action, not only who accounts for it afterward |
| Bureaucracy | How actions are controlled through rules | Delay is not contestability; a long approval chain with no empowered challenger has a high deficit |
| Concept | What it primarily concerns | What the |
|---|---|---|
| Organizational silence ( | Why actors withhold voice – including through structural and climate barriers, not merely individual choice | A structural, designed interruption of the decision chain, independent of whether any individual is willing to speak |
| Voice/psychological safety ( | Whether members can and will raise concerns | Voice may be exercised without power to alter an output already shifted into an executable register; the deficit persists even under high psychological safety |
| Organizational visibility/transparency | What information becomes visible, legible or safely reportable within the organization | A visible decision can still be uncontestable; visibility is not contestability |
| Automation bias ( | Why individuals over-trust system outputs | The deficit can exist even where no one is biased, because no structural contestation point was designed |
| Human oversight ( | Whether a human is involved and in what role – including authority commensurate with control | Presence is not capacity: a human may be involved yet hold no authority to suspend or revise before execution |
| Contestable | Whether a decision can be questioned, across the system lifecycle | The organizational loss of effective contestability specifically before execution, when challenge can still prevent embedding |
| Algorithmic accountability ( | Who must answer and how, before a forum – including ex ante duties | Whether anyone could still intervene before action, not only who accounts for it afterward |
| Bureaucracy | How actions are controlled through rules | Delay is not contestability; a long approval chain with no empowered challenger has a high deficit |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.