50/100

Moderate evidence

Cite
More
EmailLinkedInX

Overloaded workers lean harder on AI agents, a survey of 292 employees finds

Study title: Navigating digital strain: The role of overload, trust, and agentic AI dependence in predicting employee adverse outcomes

Evidence type
Survey
Sample
n = 292
Sources
1 sourcePeer-reviewed
Published
2 min read

Study title: Navigating digital strain: The role of overload, trust, and agentic AI dependence in predicting employee adverse outcomes

A peer-reviewed survey of 292 employees found that information overload, system overload and communication overload were each positively associated with dependence on AI tools that plan, coordinate or carry out work tasks with some autonomy.

Read the study abstract

Agentic artificial intelligence (AI) systems are increasingly deployed in organizations to autonomously plan, coordinate, and execute work tasks. Yet, the behavioral and organizational consequences of such systems remain underexplored. Drawing on Technostress Theory, this study conceptualizes agentic AI dependence as perceived functional work dependence on AI tools with agentic or semi-agentic workplace functions. Using survey data from 292 employees, the research examines how information overload, system overlo…

Navigating digital strain: The role of overload, trust, and agentic AI dependence in predicting employee adverse outcomes
"Construction workers at a site on Södra Hamngatan, Gothenburg" by W.carter, CC BY-SA 4.0.

The problem

In a survey of 292 employees, information, system and communication overload were all positively associated with dependence on agentic AI tools at work, suggesting reliance tracks workplace strain more than enthusiasm for the tools.

In brief

The authors read this as a sign that reliance on agentic AI is tied more closely to environmental strain than to enthusiasm for the technology. They frame dependence as both a way of coping and a risk, and they extend Technostress Theory to cover behavioural adaptation as well as strain.

Two system and trust findings complicate the picture. Flexibility was unexpectedly negatively associated with dependence, which the authors say may mean customizable systems help preserve human agency. Trust in agentic AI weakened the link between information overload and dependence.

Main points

  1. The study surveyed 292 employees about agentic AI tools used for autonomous planning, coordination and execution of work tasks.
  2. Information overload, system overload and communication overload were all positively associated with agentic AI dependence.
  3. Dependence was defined as perceived functional work dependence on AI tools with agentic or semi-agentic workplace functions.
  4. Flexibility of the system unexpectedly showed a negative association with dependence.
  5. Trust in agentic AI weakened the positive link between information overload and dependence.
  6. Trust strengthened the positive association between communication overload and high integrability.
  7. The authors recommend that organizations manage digital overload, calibrate AI delegation and design systems that enhance human autonomy.

The full story

The research draws on Technostress Theory and tests how three forms of digital overload relate to dependence on agentic AI in the workplace. It also examines system attributes, namely reliability, flexibility and integrability, with trust in agentic AI as a moderating factor.

Overload of all three kinds went with greater dependence. The authors interpret this as reliance reflecting environmental strain rather than system enthusiasm, and they describe dependence as a dual coping-risk mechanism.

Customizable systems were associated with lower dependence, which the authors suggest could reflect greater preservation of human agency. Trust moderated the associations in mixed ways, strengthening one link and weakening another.

For practice, the authors call for organizations to manage digital overload, calibrate how much work is delegated to AI, and design systems that support rather than replace human autonomy.

Sources

1 on file
  1. Acta Psychologicadoi:10.1016/j.actpsy.2026.107956
    Peer-reviewed2026-11-12

Talk it over

Three questions to start on, two more if you have time
The survey of 292 employees linked information, system and communication overload to dependence on agentic AI tools. How would you tell whether overload drives the dependence or the reverse?

Start with a question

Overloaded workers lean harder on AI agents, a survey of 292 employees findshttps://truvace.com/item/navigating-digital-strain-the-role-of-overload-trust-and-agentic-ai-dependence-i

  1. The survey of 292 employees linked information, system and communication overload to dependence on agentic AI tools. How would you tell whether overload drives the dependence or the reverse?

    Answer in the thread
  2. The authors call dependence both a way of coping and a risk. How do you weigh the relief an overloaded worker gets from these tools against the risk of relying on them?

    Answer in the thread
  3. Flexibility was unexpectedly negatively associated with dependence, and trust weakened the link between information overload and dependence. What would you want an employer to change first, and why?

    Answer in the thread
If you have more time
  • The authors recommend that organizations calibrate AI delegation. What would that look like in a workplace you know, and who should decide how much work is handed over?

    Answer in the thread
  • The 292 employees reported their own dependence. What other evidence, such as task records or interviews, could show whether the reliance is a problem?

    Answer in the thread

What readers say