TRV-2026-2142Certified recordPeer-reviewed

Navigating personal control: AI augmentation and its impact on decision-making

As Artificial Intelligence (AI) becomes increasingly embedded in organisational decision-making, it fundamentally reshapes workers' personal control over their tasks and outcomes. While, in contrast to AI automation, AI augmentation is typically promoted as a way to preserve human agency and personal control, empirical evidence regarding its psychological consequences remains fragmented. This study adopts Control Theory to examine how varying degrees of AI involvement, conceptualised as a continuous gradient rat…

Labor · The Trace — both readings · certified 2026-10-22 · v1 · article view · machine-readable

Current reading — gain

Low and medium levels of AI support left participants' personal control comparable to working without AI, so limited augmentation did not measurably cost control in this task.

Current reading — problem

High AI involvement significantly reduced personal control by lowering felt responsibility, producing control loss comparable to full automation even when humans formally retained final decision authority.

What this doesn’t fix

The evidence comes from an online within-subject experiment using a single bin-packing optimisation task, so it is a controlled setting rather than observation of real organisational decision-making.

Evidence

Reader signal

How should this claim be treated?

Cite this record

Truvace Impact Record TRV-2026-2142, v1: “Navigating personal control: AI augmentation and its impact on decision-making.” Truvace, 2026-10-22. /record/TRV-2026-2142 (accessed at citation time). sha256 3fbd449d8e204ce7…

Calibration history

Every change to this record since certification, in the open. None yet — the reading has held since it entered the record.

  1. Certifiedv13fbd449d8e20…

    Certified into the record

Verify this record
How to verify without trusting this page

Fetch the canonical text of any version from /api/record/TRV-2026-2142 and hash it yourself — for example shasum -a 256 on the saved canonical field. The result must equal content_hash, and each version’s text ends with prev:followed by the prior version’s hash (version 1 chains to 64 zeros). If a single character of any version had been altered since certification, the chain would not reproduce.