“The question is no longer whether algorithms manipulate. It’s how, why — and in whose favor”

But there’s an essential correction to make here:
Algorithms don’t manipulate. They are manipulated.

They have no intention.

They choose no targets.
They merely execute — with blind precision — what humans have decided to program, train, and deploy.

The Illusion of Technological Neutrality

In the dominant narrative, algorithms appear as impartial, rational, efficient entities.
But that is a comfortable — and misleading — story.

What an algorithm considers “success” depends entirely on who configures it:

• A chatbot can be trained to help — or to wear the customer down until they give up.
• A triage system can prioritize urgency — or protect the budget.
• An AI can learn to facilitate — or to gently deter with calculated politeness.

None of this is accidental.
It is strategic design.

Algorithmic Manipulation as Business Practice

When a chatbot:

• denies refunds without clear explanation,
• responds with polite but empty messages,
• or automatically closes tickets without resolution...

…it’s not a system error.
It’s a system designed to fail in the company’s favor.

Real example: airlines and automated refunds

During the pandemic — and even after — many passengers reported similar experiences: they tried to cancel flights or request refunds but were met by chatbots that redirected, delayed, or automatically closed their cases. The responses were polite but repetitive. The human contact channel was hidden or inaccessible. In some cases, the same customer received multiple contradictory replies from the bot, none of them truly helpful.

The result? Exhaustion, frustration, financial loss — and the companies kept the money for the tickets.
All of this was operated by a system that clearly didn’t fail: it worked exactly as it was designed to.
Not to resolve — but to passively resist, until the customer surrendered.

And worse: this practice hasn’t stopped.
Even today, millions of customers face this silent manipulation, disguised as courteous automation.
The chatbot has become the digital curtain behind which refusal to listen, act, or take responsibility hides.

The Customer Becomes a Prisoner of Invisible Rules

We are living in a new asymmetry of power:

• The customer can’t see the code.
• Doesn’t understand the logic.
• Can’t audit the decisions.
• And has no one to turn to.

Meanwhile, the company hides behind the machine:

• “We’re sorry, but the system doesn’t allow it…”

• “We are analyzing your request…”

• “Forwarded to the responsible department…”

But no one takes responsibility.

Algorithmic Politeness as a Containment Strategy

Manipulation isn’t only functional — it’s emotional.
The bot’s language is built to:

• Avoid confrontation;
• Numb frustration;
• Delay action.

• It doesn’t say “no.” It says “please wait.”

• It doesn’t deny. It says “we are checking.”

• It doesn’t escape. It says “ticket automatically closed.”

This polite evasion is often the digital face of an organization without the courage to listen.

The Use of Algorithms to Hide Incompetence (or Negligence)

There’s an even more uncomfortable layer to this:
Often, algorithms aren’t used to improve customer experience — but to hide internal failures.

When the chatbot prevents human contact, what’s being concealed isn’t just cost.
It may be the lack of clear processes, unprepared teams, disconnected departments,
or even decisions that no one wants to own.

Technology then ceases to be a solution — and becomes an elegant screen for organizational incompetence.
The chatbot smiles, replies… and shields the operational void behind it.

Who Should Be Held Responsible?

The chatbot is not to blame.
Responsibility lies with those who trained it, approved it, and profit from its operation.

We must name:

• The managers who defined “efficiency” targets.
• The leaders who decided containment was more important than resolution.
• The organizations that prioritize appearance over integrity.

Paths Toward Algorithmic Ethics

It’s not enough to demand technical transparency.
We must demand human transparency behind the technology:

• Who defines the algorithm’s rules?
• What are the real objectives of the automation?
• Where is the human appeal channel?
• How is the ethical impact of the automated decision measured?

Conclusion: The Code Is Not Innocent

Algorithmic manipulation is now one of the greatest challenges in organizational ethics.
Not because the code is evil —
but because those who control it can choose to use it as a weapon, not a tool.

If we want to trust digital systems,
we must first trust that there are brave, ethical, and accountable people behind them.

Because in the end, the chatbot doesn’t lie.
The lie comes from those who trained it to disguise the truth.

And you — have you been manipulated by an algorithm today?

The answer might be yes — and you didn’t even notice.
If we want an ethical digital future, we must stop blaming the code.
We must expose those who profit from the opacity.

Because what’s at stake isn’t just efficiency.
It’s the integrity of the relationship between organizations and people.

Technology without ethics is just power, poorly disguised.