How AI Over-reliance Could Lower Employee Engagement
By Marvin Heffernan
PO, like millions of others, use AI increasingly. Indeed, history repeatedly shows that new tools can liberate us all in positive ways.
However, there is a worrying AI trend that is gnawing at the underbelly of organisational capability, so read on lest you inadvertently dumb-down your talent base……..
A recent MIT study revealed students who used ChatGPT to write essays showed significantly lower brain engagement, with weaker neural connectivity, even after switching back to human-only writing.
Insights into neural activity and meaningful work are not new. A few years ago, business psychologist Dan Cable released Alive at Work, where he explained how our ‘seeking system’, the part of the brain wired to explore, learn, and feel alive, must be stimulated if employees are to remain engaged. When the seeking system is muted, whether by bureaucracy or by brain-numbing AI, motivation and engagement evaporate.
Over-reliance on AI appears to amplify that same dynamic at scale, potentially leading to diminished cognitive skills widespread disengagement (Time, New Yorker).
From Engagement to Erosion
As countless Gallup studies show, employee engagement thrives on autonomy, mastery, and purpose. People want to feel their work matters, that they are using their judgment, and that they are growing.
Yet early evidence suggests AI deployments may undermine these very drivers. A 2024 MIT Sloan study found that while employees welcomed tools that accelerated routine drafting or summarising, many reported a creeping sense of disengagement when their role was reduced to “editing machine output” rather than exercising independent thought.
Similarly, research published in Nature Human Behaviour (2023) showed that over-reliance on algorithmic outputs diminishes confidence in one’s own reasoning, creating learned dependence and lowering creative problem-solving scores.
When engagement and creativity slip, the signs are subtle: fewer contributors in meetings, decisions deferred “because the model says so,” and a decline in constructive debate, fostering increased passivity.
Instead of knowledge workers, you have knowledge editors. Instead of a culture of critical thinking and discernment, you have a culture of rubber-stamping.
Cognitive Devolution: A Real Risk
To dismiss these emerging concerns as simple technology resistance would be folly. After all, intelligence is a muscle that strengthens through use.
Psychologists warn of the automation paradox: the more we automate, the more skilled humans must be to step in when systems fail. Pilots who spend long periods on autopilot, for instance, experience a measurable decline in manual flying ability, a finding well documented by the FAA.
A workforce fed on AI, endless half-formed drafts, templated insights, or shallow analysis, risks losing the ability to separate signal from noise. That is not only a cultural risk but a commercial one as operational effectiveness dulls, and customers quickly defect due to lack of authentic service.
Overcoming Slop Through Substance
Organisations can take deliberate steps to ensure AI augments rather than replaces human judgment, creating an environment where employees feel empowered by AI tools rather than diminished by them:
- Reframe AI as collaborator, not crutch. Encourage employees to treat AI as a sparring partner, a way to test ideas, explore perspectives, and accelerate thinking, rather than as a definitive source of truth. This reflects Harvard Business Review’s emphasis on “centaur” models of work, where human and machine thinking are interwoven.
- Invest in cognitive stretching. Just as physical health requires exercise, so too does mental agility. Provide opportunities for employees to problem-solve without AI, through workshops, scenario planning, or job rotations. A McKinsey report highlighted adaptability and complex problem-solving as the most in-demand skills for the coming decade.
- Redesign engagement metrics. Traditional employee surveys often fail to capture whether staff feel cognitively stimulated. Incorporating expectations around autonomy, challenge, and growth, as outlined in self-determination theory, can help leaders detect early signs of stagnation.
- Guard against slop in outputs. Set standards for AI-generated work. If employees know that quality checks will scrutinise nuance, accuracy, and originality, they are more likely to apply judgment rather than blindly accept machine drafts.
From Rhetoric to Reality
Many companies boast of their “AI transformation” without considering its long-term unintended consequences.
If workers become entirely dependent on AI, we may usher in an age not of progress but of deskilling.
The choice, now as ever, lies in how we wield the tools. Companies that treat AI as an enabler of cognitive vitality will nurture rather than malnourish the skills that drive lasting outcomes.
———————-
If you want to improve employee engagement using a company that still values the human, contact us here.