AI: the ghost in the corporate machine

Every Monday morning chief executives across the country hit send on carefully crafted, all-staff emails designed to rally the troops and align corporate strategy. They review the subsequent analytics dashboard, spot a healthy 85% open rate, and breathe a sigh of relief.

 

But in 2026, those analytics are lying. While leadership assumes its voice is landing precisely as intended, the gap between delivery and understanding is only getting wider. The modern workforce is suffering from an acute "attention recession". Faced with an overwhelming deluge of corporate information, employees have quietly abdicated their role as readers and handed the job over to AI summarisers.

 

 

The “Employee Attention Recession”

Recent independent research reveals that a staggering 83% of employees feel bombarded by too much internal content, with more than a third admitting they simply cannot keep up. In response to this volume fatigue, a mere 12% of professionals bother to read internal communications in full, making reading the complete text the least common active response to a company update.

 

Instead, the workforce has modernised its filtering systems. Some 88% of employees have used AI to condense internal messaging, and more than half now do so as a matter of regular routine.

 

AI-generated summaries have bypassed intranets, leadership emails, and corporate chat channels to become the number-one discovery method for company news. Nearly 28% of employees now encounter major corporate updates through an automated summary first, making algorithms the primary gatekeepers of corporate information.

 

 

The trust vs distortion tensions

The threat here is not the technology itself. AI tools are exceptionally efficient at reducing cognitive load, and in an era of persistent information overload, that function has genuine value. The problem lives in a structural disconnect around accuracy, and more specifically, around who gets to define what accurate actually means.

 

A full 95% of employees trust their AI tools to faithfully capture the key points of any internal message. At the same time, 92% of the internal communications professionals responsible for crafting those messages harbour serious concerns that automated summaries distort core meaning, tone, and emotional register. Both groups believe they are working with the truth. Only one of them is right.

 

The reason becomes clear the moment you examine what compression actually does to language. An algorithm reducing a nuanced restructuring memo to three bullet points will handle functional content with ease, timelines, actions and figures. What gets stripped away is everything harder to quantify: the empathy, the strategic context, the careful acknowledgement of uncertainty, and the subtle cultural reassurance that separates a workforce that feels informed from one that feels abandoned.

 

 

Subtraction is a strategic necessity

The oversight gap makes this worse. While 98.6% of communication managers know their workforces are using AI to screen out the noise, only 33% of businesses have formal policies or monitoring systems governing how content gets reinterpreted off-platform. Most remain stuck in a reactive loop, focused on using AI to generate more content, rather than managing how that content gets digested.

 

This represents an urgent design problem for corporate communication. For years, the default response to poor alignment has been addition: more emails, more channels, more intranet pages, more video updates, under the impression that more accessibility translates to greater clarity. In an attention-constrained environment, that strategy backfires, driving burnout.

 

Surviving the automated summary requires a different instinct entirely. Leaders need to treat employee attention as the finite resource it actually is. That means auditing communication channels, sequencing major announcements with discipline, and developing the organisational confidence to say no when secondary stakeholders push for all-staff coverage.

 

Getting corporate messaging right comes down to trust. The leaders who cut through aren’t producing more content or broadcasting on more channels; they’re saying things that hold up, feel honest, and still mean something after passing through five layers of management and a town hall Q&A.

 

When people genuinely hear their leader, that’s never an accident. That comes from language built to last, not only to land.

 

---

 

David Bowman is Product Director at Fresh, an Advania company

 

Main image courtesy of iStockPhoto.com and designer491