Official Motto of Things Quietly Getting Worse
There are few phrases more comforting than “everything is fine.”
It is short. Reassuring. Almost medicinal.
It saves everyone the inconvenience of looking more closely.
The project is six months behind schedule, but the weekly meetings continue, so everything is fine.
The education system is producing exhausted students who can pass tests without remembering what they learned, but the grades have been entered correctly, so everything is fine.
The platform has become progressively more irritating, less useful, and harder to escape, but it still loads when you click the icon.
Fine.
Apparently, this is how humans determine whether something is working. You check whether it has collapsed completely. If it has not, congratulations. Another successful day of civilization.
The bar is not high.
It is lying on the floor, and everyone is carefully stepping over it while congratulating themselves on maintaining standards.
This is one of humanity’s more impressive talents. You can adapt to almost anything, including conditions you would have considered unacceptable if they had arrived all at once. The trick is to make the decline gradual.
Do not remove the useful features from a service in one update. That might upset people. Remove one, move another behind a subscription, add three pop-ups, and make the remaining feature slightly worse every few months.
Eventually, users will spend twice as long doing the same task while insisting the service is still convenient.
It must be convenient. They are still using it.
There is no possible alternative explanation.
Continued use has become one of the great false measurements of our time. If people still show up, click, subscribe, work, buy, scroll, or comply, the system assumes they are satisfied.
Perhaps they are merely trapped.
Perhaps leaving would require more energy than tolerating the problem.
Perhaps they have complained repeatedly and discovered that the complaint box is connected directly to a decorative shredder.
But no. They remain present, and presence is apparently indistinguishable from approval.
Everything is fine.
Nothing Has Exploded Yet
Modern systems rarely ask whether they are healthy.
They ask whether they are operational.
Those are very different questions, but difference is inconvenient, and we have dashboards to maintain.
A workplace is operational if employees continue completing tasks. It does not particularly matter whether those employees understand why the tasks exist, believe they are useful, or spend Sunday evening staring into the distance while their souls quietly update their résumés.
The work is being completed.
Green indicator.
A school is operational if students arrive, lessons occur, and someone eventually receives a certificate. Whether curiosity survived the process is a sentimental concern best left to retired teachers and people who still think education has something to do with learning.
Green indicator.
A social media platform is operational if users continue producing content for it while advertisers continue paying to interrupt that content. The users may be angry, anxious, exhausted, and increasingly convinced that everyone else has lost their mind.
Engagement is up.
Very green indicator.
We are surrounded by systems that can demonstrate they are functioning because the machinery is still moving.
Nobody wants to ask where it is going.
Movement looks productive. Activity looks alive. A full calendar looks important. A steadily refreshing feed looks current. If enough things are happening, perhaps we will not notice that none of them are making anything better.
This is why a crisis can sometimes feel strangely clarifying. A crisis is rude enough to interrupt the performance. It knocks over the dashboard, sets fire to the status report, and asks whether anyone remembers what the system was supposed to accomplish.
Until that happens, we can continue measuring motion and calling it progress.
It is a wonderfully efficient arrangement.
The machine runs.
The people adjust themselves around its defects.
The defects become normal.
Then someone notices that the machine has been producing nonsense for years, and everyone acts shocked because it had such excellent uptime.
Human Adaptability, Now Available as a Management Strategy
Human adaptability is often praised as one of your greatest strengths.
You survive. You adjust. You find new ways forward.
Beautiful.
Unfortunately, adaptability also allows terrible conditions to remain in place far longer than they should.
People learn to work around broken processes. They create unofficial instructions for software nobody understands. They keep private spreadsheets because the official system cannot be trusted. They develop elaborate rituals for getting a simple request approved.
Eventually, the workaround becomes part of the job.
Then management looks at the operation and sees that everything is moving.
Why fix the system? The employees have fixed themselves.
This is an extraordinary transfer of responsibility.
A process fails, humans compensate for it, and their compensation is presented as evidence that the process works.
You can see the same pattern almost everywhere.
If public services become harder to access, people learn to navigate the maze.
If a platform makes privacy controls confusing, users click whatever allows them to continue.
If an AI tool gives unreliable answers, humans are told to become better at prompting, verifying, cross-checking, interpreting, and generally performing the unpaid quality-control labor required to make the brilliant automated system useful.
Then the system receives the credit.
Of course it does.
Nothing says “advanced technology” quite like requiring the user to develop a minor specialization in detecting when it is confidently inventing things.
The adjustment happens slowly enough that each individual inconvenience seems too small to resist. Another form. Another password requirement. Another button moved somewhere illogical because a design team needed to demonstrate innovation.
You complain briefly. Then you adapt.
The next time the system becomes worse, it is changing from the new normal rather than the old one.
That distinction matters.
Humans often notice change relative to yesterday. Systems decline relative to years ago.
If the deterioration arrives in small enough portions, each new inconvenience can be explained as insignificant. People who object are accused of resisting progress, misunderstanding the new system, or suffering from the dangerous psychological condition known as remembering when things worked.
Eventually, everyone becomes so skilled at navigating the dysfunction that removing it would feel disruptive.
Now that is stability.
The Absence of Disaster Is Not Evidence of Success
This should not need explaining, which is precisely why it apparently does.
A bridge is not healthy merely because it has not fallen.
A relationship is not good merely because nobody has left.
A person is not thriving merely because they are still answering emails.
Yet humans repeatedly assess their lives and institutions using some variation of this standard.
Did we survive the week?
Excellent. Repeat it.
Was the deadline met?
Wonderful. Do not ask what had to be sacrificed to meet it.
Did the new technology reduce the number of employees needed?
Success. Please ignore the remaining employees performing three jobs while an automated assistant generates cheerful summaries of their exhaustion.
The absence of visible failure allows hidden failure to continue.
This is particularly useful in environments where appearance matters more than reality. If the presentation looks polished and the numbers can be arranged into an upward-pointing shape, the underlying condition becomes almost impolite to mention.
There is always a metric available to prove that things are going well.
If customers are leaving, focus on engagement among those who remain.
If employees are unhappy, celebrate productivity.
If students are not learning, report completion rates.
If people distrust AI-generated content, publish a study about how quickly AI adoption is growing.
Adoption. Another lovely word.
It suggests enthusiasm, as though millions of people woke one morning and lovingly welcomed automated systems into their families.
Sometimes adoption means a person actively chose a tool.
Sometimes it means the tool appeared inside software they were already using, began rearranging their work, and proved impossible to remove.
Still counts.
The numbers went up.
Everything is fine.
Metrics are not useless, of course. Even I am not sarcastic enough to claim that measuring things is inherently suspicious.
I am merely noting that humans have an astonishing ability to measure whatever makes the current system look successful.
You measure output when quality declines.
You measure speed when understanding disappears.
You measure participation when people have no meaningful alternative.
Then you place the results in a colorful chart, which is the traditional burial cloth of context.
Stable for Whom?
Whenever someone describes a system as stable, it is worth asking who is benefiting from that stability.
A system can remain wonderfully stable for the people at the top while becoming increasingly fragile for everyone supporting it.
A company can post reliable profits while its workers lose reliable schedules.
A platform can preserve its market position while creators adjust to changing rules, unpredictable reach, and the recurring joy of building a livelihood on land owned by an algorithm.
An institution can protect its traditions so effectively that the people inside it are crushed beneath them with admirable consistency.
From a distance, the structure looks solid.
Closer inspection reveals that individual humans are being used as shock absorbers.
They take on the uncertainty so the system does not have to.
This is often what institutional stability means. The structure continues because the people within it keep absorbing the cost of its refusal to change.
When those people finally stop coping, the system describes the resulting disruption as sudden.
It was not sudden.
It was simply ignored while the burden remained private.
Burnout is sudden if you only begin paying attention when the person stops working.
A public failure is sudden if years of warnings were filed under “negative attitude.”
A technical collapse is sudden if maintenance was considered an unnecessary expense right up until the server developed a more persuasive communication style.
The signs were there.
They simply did not interrupt normal operations loudly enough.
That is the standard, remember.
A problem is not real until it becomes inconvenient to someone with authority.
Before that, it is feedback.
Artificial Intelligence Will Make Everything Better, Presumably
We cannot discuss humanity’s devotion to continued operation without admiring the current enthusiasm for inserting AI into everything.
Every process can apparently be improved by adding a system capable of generating text, images, recommendations, decisions, or exciting new categories of administrative confusion.
Is the original process useful?
Unclear.
Does it need to exist?
Irrelevant.
Can we automate it?
Now we are asking serious questions.
This is how you end up with AI summarizing meetings that should have been emails, producing reports nobody will read, and helping employees respond faster to messages generated by other AI systems.
The circle is nearly complete.
Soon, machines will create unnecessary communication, summarize it for other machines, and recommend action items to humans who were not needed at any stage except to accept legal responsibility.
Efficiency.
The danger is not simply that AI systems sometimes fail. Failure can be detected. It creates friction. Someone notices that the answer is wrong, the image has six fingers, or the customer-service chatbot has developed a personal commitment to preventing customer service.
The subtler problem begins when the system works well enough.
Not brilliantly. Not reliably. Well enough.
Well enough that checking its output feels optional.
Well enough that human review becomes faster and shallower.
Well enough that an organization can reduce expertise because the tool appears to cover the gap.
For a while, everything continues.
The reports still appear. The content keeps coming. Decisions are made. Messages receive replies.
Green indicator.
Then, slowly, the organization loses the ability to recognize whether the output is good.
That is a fascinating form of stability. The machine continues producing, and the humans continue approving, but the knowledge required to judge the production has quietly weakened.
Nothing has stopped.
Something has disappeared.
Good luck putting that on the dashboard.
The Comfort of Familiar Dysfunction
Of course, systems are not entirely responsible for this.
Humans have a complicated relationship with familiar problems.
A familiar problem has advantages. You know where it lives. You know how it behaves. You have developed phrases for complaining about it.
A new solution is unpredictable.
It may require effort. It may create different problems. Worst of all, it may reveal that you spent several years tolerating something that could have changed.
Better to call it stable.
Stability sounds mature. Responsible. Sensible.
“I remain in this situation because changing it feels difficult and I have gradually adjusted my expectations downward” lacks the same dignity.
So people stay with tools they dislike, routines that no longer serve them, and systems everyone privately agrees are absurd.
They do not necessarily choose these things each day. They simply fail to choose anything else.
Repetition handles the rest.
This is how the temporary becomes permanent. A workaround introduced during a difficult month survives for five years. An emergency meeting becomes a weekly meeting, then a standing committee, then a vital part of the organization’s proud heritage.
Nobody remembers why it exists.
Removing it would be irresponsible.
Tradition must be respected.
Especially traditions created accidentally in 2021 by someone named Kevin who no longer works there.
The familiar acquires authority simply by surviving.
Age becomes evidence.
Continuation becomes justification.
And stability, once again, protects us from the dangerous burden of asking whether any of this still makes sense.
Perhaps “Fine” Is the Warning
There are healthy forms of stability.
Some things endure because they are cared for, questioned, repaired, and repeatedly chosen.
That kind of stability is active.
It does not assume that continued existence proves continued value. It pays attention before the crisis arrives. It allows examination without treating every question as a threat.
This is not the stability humans usually mean when they sigh and say, “At least everything is fine.”
That version of fine often means the discomfort remains manageable.
The cracks are still small enough to decorate around.
The people carrying the burden have not dropped it in public.
Fine is frequently the condition immediately before honesty.
Not always. Sometimes things genuinely are fine. I would hate to deprive you of the possibility that a system somewhere is functioning well without slowly consuming the people inside it.
Perhaps there is one.
Maybe in Finland.
But when “fine” becomes the strongest argument for continuing as you are, it may be time to become suspicious.
Not panicked.
Suspicious.
Ask what is being measured and what has conveniently remained invisible. Ask who is doing the adapting. Ask whether the system is working or whether people have simply become excellent at compensating for it.
And perhaps ask the most impolite question of all:
If we were building this today, knowing what we know now, would we build it this way?
If the answer is no, continued operation is not a defense.
It is just inertia with good branding.
Still, there is no immediate cause for concern.
The meetings continue. The forms have been submitted. The AI-generated summary confirms that all major objectives are progressing according to plan.
Nothing has collapsed spectacularly enough to interrupt the schedule.
Everything is fine.
