AI Use at Work Lowers Manager Trust in Staff

Managers should expect AI-assisted staff to be judged less favourably unless a clear company policy is in place.

Updated on Nov. 6, 2026 in Remote Work

Gouache-painted editorial illustration of a cream envelope tucked under a tilted desk plane against a muted sage field, evoking workplace trust.
Managers rate employees who use large language models at work less favourably, according to a six-study paper in The Leadership Quarterly. AI Illustration. Upload story photo >

Live Poll

Would you trust a coworker less knowing they used ChatGPT for their work?

Employees who use large language models (LLMs) at work face lower trust from their managers, according to a six-study paper in The Leadership Quarterly. The effect appears across international workplaces that have no clearly stated AI policy, and it is weaker where companies explicitly encourage LLM use.

Why it matters

For operators, the finding means AI adoption carries a reputational cost that rides on the manager relationship, not just on output quality. Lower supervisory trust was tied to weaker perceived performance and reduced promotability, so teams deploying AI tools without clear policies may see unintended effects on how their staff are evaluated.

Across the experiments, supervisor trust fell from 68% for independently completed work to 24% for AI-generated output. The paper is built on six studies, five of them incentivised economic experiments and one a field study of academic supervisors.

The players

The Leadership Quarterly

A peer-reviewed academic journal focused on leadership research, publishing scholarly work used by management theorists and HR practitioners.

The details

Supervisors in the experiments were asked whether they would accept an employee's account of how long a task took or independently verify the information. The trust penalty persisted when organisations prohibited AI use or had no clearly stated policy, but it was no longer statistically significant when companies explicitly encouraged LLM use. The field study separately measured how students' broader LLM use related to supervisors' reported trust in them.

Timeline

  1. Article on the research published 6 November 2026.

Market Landscape

The study adds an empirical data point to a growing body of management research on how AI use is perceived by supervisors. Its findings suggest that explicit organisational AI policy may shape how much of a trust penalty employees face.

Managers should expect that staff using AI tools may be judged less favourably unless the company sets a clear policy on AI use. Operators should review how AI use is communicated and documented as part of performance and promotion processes.

The takeaway

The strongest lever in the study is an explicit organisational stance on AI use, which removed the statistically significant trust penalty when companies encouraged it. Operators should draft and circulate a written LLM policy and track supervisor feedback on AI-assisted work as a near-term metric.

Further reading

For more on how AI tools are reshaping how distributed teams are supervised, see Remote Work.

Source note: This article includes information reported by HR Katha.

Live Poll

Would you trust a coworker less knowing they used ChatGPT for their work?