She had stopped counting the letters.
Not because they were few—but because they were too many.
On her desk, stacked in quiet, clinical symmetry, were referral forms to psychiatric hospitals. Each one carried the same sterile language: adjustment disorder, major depressive episode, somatic symptom disorder, burnout. Each one bore her signature.
She was a psychological counselor contracted to an IT company that liked to describe itself as “pre-AI legacy transitioning to intelligent infrastructure.” It sounded elegant. It meant they were late.
And lateness, in this industry, had consequences.
The company’s internal dashboard told the story more honestly than any executive memo.
A red line—“AI capability gap”—trended upward.
A green line—“operational headcount efficiency”—was expected to follow.
It hadn’t yet.
So the board made a decision.
Restructuring.
⸻
Outside her office, nothing looked different. Engineers still argued about model drift and hallucination rates. Product managers still spoke in timelines they didn’t believe in.
But something had shifted.
People were getting sick.
Not all at once. Not dramatically. But steadily.
Headaches that wouldn’t resolve.
Stomach pain with no organic cause.
Insomnia that stretched into weeks.
A senior developer who reported hearing whispers in server rooms that were, objectively, empty.
She evaluated them all.
Carefully. Methodically.
She had trained for this. She believed in diagnostic precision. She rejected the lazy shortcuts—no, not every exhausted engineer was “burned out,” not every anxious manager was “depressed.”
But the pattern was undeniable.
⸻
The company wasn’t unique.
Across the industry, layoffs were accelerating under the banner of AI transformation. Tens of thousands of workers had already been cut in early 2026 alone, with companies restructuring aggressively to redirect resources into AI systems and infrastructure .
Officially, it was about efficiency.
Unofficially, it was about survival.
And somewhere between those two explanations, people began to break.
⸻
She noticed something else.
Not everyone broke the same way.
A junior engineer came in—pale, shaking, unable to sleep. He had tied his entire identity to the company. No savings. No alternative skills. No plan B.
“I just need to survive this restructuring,” he told her. “If I get cut, it’s over.”
She referred him.
Another case: a mid-career architect.
“I saw this coming,” the architect said calmly. “I’ve been taking courses in distributed AI systems. I’ve got three contacts at startups. If I stay, fine. If not, I move.”
He had insomnia too. Stress, certainly.
But he wasn’t collapsing.
She did not refer him.
⸻
The research supported what she was seeing.
Workplace restructuring—especially downsizing—was linked to increased mental health issues and sickness absence, particularly when combined with uncertainty and negative workplace behaviors .
But research didn’t capture the nuance she saw every day.
Two people under the same pressure could diverge completely.
⸻
Management, of course, had its own interpretation.
They never said it directly.
But in closed meetings, the language was clear enough:
“High-risk employees.”
“Reduced resilience profiles.”
“Medical attrition pathways.”
If enough people exited through medical channels, the restructuring numbers looked softer. Cleaner.
More humane.
⸻
One afternoon, she was called into a strategy session.
A slide deck glowed on the wall.
“Given current trends,” an executive said, “we expect increased psychological leave. We should ensure our support systems can scale.”
Support systems.
She almost laughed.
⸻
That evening, she reviewed her cases again.
She noticed something she hadn’t allowed herself to articulate before.
Weakness was not what they thought it was.
It wasn’t emotional sensitivity.
It wasn’t stress.
It wasn’t even illness.
Weakness, she realized, was the absence of alternatives.
The people who collapsed were those who had nowhere else to stand.
No savings.
No transferable skills.
No network.
No imagination of a future outside the company.
They weren’t fragile.
They were trapped.
⸻
She opened a blank referral form.
Paused.
Closed it again.
For the first time, she began adding something new to her consultations.
Not just diagnoses.
Questions.
“What would you do if you had to leave tomorrow?”
Most had no answer.
A few did.
And those few—no matter how exhausted, anxious, or sleepless—did not fall first.
⸻
Outside, the company continued its transformation.
Budgets shifted. Teams dissolved. New AI hires arrived even as old departments disappeared.
The paradox played out everywhere: investment rising, headcount shrinking, uncertainty spreading .
Inside her office, the stack of letters grew more slowly.
Not because fewer people were suffering.
But because she had stopped confusing suffering with collapse.
⸻
Weeks later, during another executive briefing, someone asked:
“Can we predict who’s at risk?”
She considered answering.
There were machine learning models now that claimed to predict burnout using behavioral data, chat patterns, performance metrics.
They were impressive.
They were also missing the point.
She said nothing.
Because the truth was too simple.
And too inconvenient.
The weakest would always fall first.
But weakness had nothing to do with how much a person could endure.
It had everything to do with whether they had somewhere else to go.
All names of people and organizations appearing in this story are pseudonyms
Comments