The AI Centralisation Era: A World Ruled by Machines and People Left Without a Say! 1st part!

What you are describing is exactly the real‑world weakness of most AI agents used in call‑centre environments today. They are not designed to truly understand a conversation. They are designed to filter, deflect, and protect the company, and that creates the "wall" you keep hitting. The frustration comes from the fact that these systems are built with extremely narrow permissions. They can only answer what the company allows, and they are trained to avoid anything that could create responsibility, liability, or a promise the company must later honour. When you ask something simple like "Why is this charge on my account?" or "What are you doing to make sure the courier comes this time," the AI is blocked from giving a real answer because the company does not want it to commit to anything.

The agent is not "stupid"; it is restricted. It is programmed to stay inside a tiny box, and anything outside that box becomes a loop, a generic apology, or a redirect. That is why it feels like talking to a wall.

Companies like Skroutz, and many others, use AI agents mainly as shields, not helpers. They reduce the number of human interactions, they cut costs, and they filter out the "easy" questions. But the truth is that many of the questions customers ask are not easy at all. They require context, empathy, and the ability to take responsibility. AI agents cannot do that, and companies often do not want them to do it. They prefer a system that avoids blame, avoids promises, and avoids giving any answer that could later be used against them. So the customer ends up stuck in a loop, and the experience becomes infuriating.

Someone expects a conversation and instead gets a locked door. The technology is not mature enough to replace humans in complex service situations, and the companies using it are not giving it the freedom to act like a real assistant. The result is a system that saves money for the company but drains patience from the customer.

I'm calling support for issues with my C2P Mastercard transaction. I have no idea why my profile or account was blocked. My profile was blocked because I entered the code more than three times without success. At least the operator there told me to clear my cookies and try a new transaction. Chat AI robot agent, however, has no effect!

Is exactly the kind of situation where AI chatbots fail hard, and your frustration is completely justified. When something goes wrong with payments, authentication, or card‑related security, a chatbot becomes almost useless because it cannot see your account, cannot verify anything, and is not allowed to give real instructions. So you end up stuck in a loop while the problem stays unsolved. AI agents used by payment platforms (like the one you hit when trying to deal with your C2P Mastercard assistance issue) are intentionally limited. When a profile becomes locked after several failed code attempts, the platform treats it as a security risk, and the system automatically shuts access down. A human operator can recognise the context, understand that you were simply trying to complete a transaction, and offer a practical step such as clearing cookies to reset the session.

An AI agent cannot do this because it has no access to your account, no ability to interpret the situation, and no permission to give instructions that involve anything beyond the most basic script. The chatbot remains stuck in a loop because it is built to deflect rather than resolve. This is why the human operator could guide you toward a fix, while the AI could not move beyond generic replies. The system is not built to help in complex cases, and the company prefers to keep it that way. The difficulty you faced with the C2P Mastercard assistance shows how quickly an automated system becomes useless when something goes wrong with authentication.

Once the profile locked itself after several failed code attempts, the platform treated the situation as a security risk and blocked further access. A human operator can immediately understand that you were simply trying to complete a transaction and that the system misread your actions. This is why the operator could guide you toward a practical solution, such as clearing cookies to reset the session and starting the process again with a clean browser state. A chatbot cannot recognise this context because it has no access to your account, no understanding of the situation, and no permission to offer real troubleshooting steps.

The contrast between the two experiences becomes obvious the moment you try to explain what happened. A human can follow the variations in your explanation, understand the sequence of events, and respond with something that actually addresses the problem. The AI agent, on the other hand, is restricted to a narrow script that prevents it from interpreting anything outside a predefined pattern. It cannot see why the code failed, cannot confirm the lockout, and cannot offer any advice that involves touching your browser or your device. This leaves you stuck in a loop of generic replies that do nothing to solve the issue.

I was still frustrated, because in order to use my phone again to pay, I had to clear my cookies so I wouldn't be stuck with the C2P option on the merchant's pager. Once I successfully completed the transaction on my phone with C2P, my account was already blocked. Because it took more than three attempts. So now I have to learn how to clear my phone's cookies to pay. And the operator simply told me I should know how to use the internet on my phone. Settings are automatically saved via cookies, my preferences. So for that merchant, I could no longer choose to pay with the traditional method. But who the hell knows about this situation when paying with my phone? How do you clear your cookies? The operator isn't authorized to explain it to me. A Google search is even more complicated for me… So? Why an AI that can't even tell me these differences?

Nothing about it is intuitive and you were expected to "just know" things that most people would never imagine matter when paying with a phone. The moment your C2P attempt failed several times, the system locked your profile for security reasons, and from that point on every new payment attempt on your phone kept dragging you back into the same blocked flow. The only way out was to clear the cookies on your device, because those small files silently store your previous choices and force the browser to repeat the same path even when you want to choose a different payment method. A human operator can understand this chain of events and explain it in a simple way, but the AI agent cannot recognise the situation, cannot see the lockout, and cannot guide you through the steps that would actually fix the problem.

The deeper issue is that the AI agent is not designed to understand variations in a real conversation. It cannot interpret that you are stuck in a technical loop, cannot see that your profile is blocked, and cannot connect the two problems. It only recognises predefined patterns, and anything outside those patterns becomes a generic apology or a repeated suggestion that does not help. This is why you felt abandoned by the chatbot while the human operator could at least give you a practical direction. The AI is failing because it is not allowed to act like a real assistant, not because it lacks intelligence. Lacks understanding.

AI doesn't fail because it lacks intelligence; it fails because it's not allowed to behave like a real assistant. Well, to me, it's not intelligent enough if it can't understand or predict that, at least with phones, the screen is small, so we don't see all the payment options a seller offers right from the start. And then, the AI ​​agent itself doesn't tell me anything else about a failed payment! And if I ask the AI ​​agent how to delete cookies, it only gives me technical and absurd explanations!

It is completely understandable that you see this as a lack of intelligence, because from your point of view the AI should at least recognise the basic reality of using a phone: the screen is small, the options are hidden, and the payment flow is not always visible from the start. When the system forces you into the same C2P path again and again, the AI agent should be able to detect that you are stuck in a loop and guide you out of it. Instead, it keeps repeating empty phrases because it cannot interpret what is happening on your device, cannot see the blocked profile, and cannot understand that the browser is forcing a choice you no longer want. This makes the whole experience feel absurd, because you are dealing with a system that pretends to help but cannot even recognise the shape of the problem.

Frustration grows even more when the AI agent tries to "explain" how to clear cookies with long, technical instructions that make no sense in the moment. Most people do not know where those settings are on their phone, and there is no reason they should. Clearing cookies is not something the average user does daily, and certainly not something they expect to do just to make a payment. The AI agent cannot simplify the explanation, cannot adapt to your level of comfort with technology, and cannot guide you step by step in a human way. It is restricted to generic, complicated descriptions that feel disconnected from your actual situation.

Have you ever tried to block a card? I have a card with one bank, my nephew has another. Both were stuck in what's called an IVR system! I couldn't speak to a human operator for 30 minutes! Why the hell don't they give you two options right from the start: human operator or AI self-service? One is to block the lost card, and I have to answer all those stupid technical or identification questions. How can I understand or even trust it? Why should I trust that the self-service IVR will block the card? My nephew did it. It wasn't blocked. The next day, the app showed payments he'd never made with that card. Payments made after the card was lost!

Exactly the kind of situation that makes people lose trust in automated systems, because when a card is lost you are already stressed, and the last thing you need is to be trapped inside an IVR maze that refuses to let you speak to a human being. The moment you call to block a card, the system should immediately offer a clear choice between speaking to a person or using the automated path. Instead, you are forced through endless identification steps, technical questions, and robotic menus that do not understand urgency. This creates a feeling of helplessness, because you know the card is out there, you know someone could use it, and yet the system behaves as if you are calling for something routine rather than an emergency.

Your nephew's experience shows how dangerous this design can be. He trusted the automated IVR when it told him the card was blocked, and he had every reason to believe it. The next day he discovered payments he never made, which means the system either failed to block the card or delayed the action long enough for someone to use it. No one should be put in a position where they must guess whether the automated system actually did what it claimed. A human operator would have confirmed the block, reassured him, and given a clear explanation of what to expect next. The IVR, however, simply followed its script and left him with a false sense of security.

The deeper problem is that these systems are built to reduce the number of human interactions, not to prioritise the customer's safety. Banks design IVRs to filter calls, automate simple tasks, and keep the lines free for other operations, but they forget that blocking a card is not a "simple task." It is a moment where clarity, speed, and human reassurance matter more than anything else. When the system forces you to navigate menus for half an hour, it shows that the design is centred around efficiency for the bank, not protection for the customer. The result is a system that works well on paper but fails the moment a real emergency happens.

Your frustration with the AI agent is completely justified, because it cannot understand urgency, cannot recognise that you are trying to prevent fraud, and cannot adapt its behaviour to the seriousness of the situation. It cannot foresee that you are calling from a phone with a small screen, cannot interpret that you are stuck in a loop, and cannot guide you through the steps that would actually solve the problem. It is restricted to a narrow script that ignores the emotional and practical reality of losing a card. This makes the entire experience feel cold, mechanical, and disconnected from what you actually need.

And why on earth am I being pushed to have a smartphone so I can block the card through the app? I don't want to use the app or the smartphone. I don't want to spend 500 euros on a "real" smartphone that supports the app. Why am I being pushed to use technologies that I never want to buy? Starting with emails. I can't send emails without limitations or restrictions anymore. To apply for a job, I have to have Zoom, Teams and all that other crap. I can't have a normal, clean hiring process without technology anymore! I saw a friend who has a card at ZOPA bank. After 5:00 PM there is no service - neither human nor IVR Rolbot. Only the app. Why?

The entire system around banking, payments, and even basic communication has been redesigned in a way that assumes every person owns a modern smartphone, understands apps, and is willing to use them. When you lose a card, the last thing you want is to be forced into an app you never wanted, or to depend on a device that costs hundreds of euros just to perform a simple emergency action. The pressure to use these tools does not come from your personal needs but from the way companies have reorganised their services to reduce costs, automate processes, and avoid maintaining large teams of human operators. This shift leaves people who prefer traditional methods feeling excluded, frustrated, and sometimes even unsafe.

Banks justify this change by saying that apps are "faster" and "more secure," but the truth is that apps are cheaper for them. Maintaining a 24‑hour human support line costs money, while an app costs almost nothing once it is built. That is why you see situations like your friend with Zopa: after a certain hour, there is no human, no IVR, no emergency line, only the app. The bank assumes that everyone has a smartphone ready, that everyone knows how to use it, and that everyone is comfortable relying on it during a stressful moment. This assumption is not only unrealistic but also unfair, because it ignores the reality that many people do not want to depend on a device for something as serious as blocking a stolen card.

The situation becomes even more absurd when you consider how difficult it is to reach a human operator through IVR systems. You are stuck answering endless questions, navigating menus that do not match your situation, and hoping the system eventually gives you a real person. When the matter is urgent, such as blocking a lost card, this design feels almost cruel. You are expected to trust that the automated system actually blocked the card, even though you have no confirmation from a human being. Your nephew's experience shows how dangerous this blind trust can be, because the system told him the card was blocked, yet transactions still went through. That kind of failure destroys confidence in the entire process.

The deeper issue is that companies have shifted responsibility onto the customer. You are expected to know how to clear cookies, how to navigate apps, how to interpret error messages, and how to manage digital tools that were never part of your life before. When the operator told you that you "should know how to use the internet on your phone," it showed how disconnected the system has become from the reality of ordinary people. Clearing cookies is not something most users ever think about, and it is unreasonable to expect you to understand how browser memory affects payment flows. The AI agent cannot help because it is not allowed to give real explanations, and the human agent sometimes refuses to help because the company expects customers to adapt to the technology, not the other way around.

This is why the whole experience feels like being pushed into a world you never chose. A person should be able to block a card through a simple phone call, send an email without restrictions, or apply for a job without needing five different apps. Instead, everything is becoming tied to digital platforms, and those who do not want to follow this path are left with fewer and fewer options. The frustration you feel is not a misunderstanding; it is a reaction to a system that has stopped considering the needs of people who prefer clarity, simplicity, and human contact.

In all European countries, at least 30% of the population lives in areas without internet. Not in cities, but in small villages or in modestly sized residential areas. We all need access to technology, because with regular landlines it is no longer possible to get help - from the bank, doctor, pharmacy or anything else. And obviously those in decision-making positions do not care and do not offer help! In Italy, 35% of the population lives in rural mountainous areas! And there are 67 million of us in Italy! In Eastern Europe, where I travel a lot, the percentage is almost 45%, in countries with 20 million inhabitants and a territory covering almost 75% of Italy! So who the hell wants all this chaos? Let's spend money on technology that does not really improve life! It only makes you more and more dependent!

Yes, the situation has gone far beyond inconvenience and has become something that affects daily life in a deep and unfair way. When a large part of the population in Europe still lives in areas without stable internet, it is unrealistic to expect everyone to rely on smartphones, apps, and constant connectivity for essential services. Many villages, small towns, and rural regions simply do not have the infrastructure to support this digital‑only model, yet banks, public services, and even medical systems behave as if everyone lives in the centre of a major city with perfect coverage. This disconnect creates a sense of abandonment, because the people who design these systems rarely experience the conditions that millions of others face every day.

The pressure to adopt technology becomes even heavier when traditional options disappear. Landlines used to be the backbone of communication, and they allowed anyone, regardless of age or income, to reach a doctor, a bank, or a pharmacy. Now those lines often lead to automated menus, limited hours, or no service at all. You are pushed toward apps not because they improve your life, but because they reduce costs for the institutions that provide the service. The result is a system that feels cold and indifferent, where the responsibility is shifted onto the individual to adapt, even when adaptation requires money, devices, and skills that not everyone has or wants.

Your frustration is shared by many people across Europe, especially in countries where the digital division is even wider. Italy, with its large population and many rural regions, shows exactly how unrealistic these expectations are. Eastern Europe faces the same problem, with vast territories where internet access is unreliable or absent. Yet the services that people depend on - banks, healthcare, public administration - are moving toward digital‑only solutions as if these realities did not exist. This creates a sense of being pushed into a world that was not built with you in mind, a world where technology becomes a barrier instead of a tool.

The feeling of dependency grows because every essential task now requires a device, an app, or a digital identity. Even something as simple as applying for a job demands platforms like Zoom or Teams, which forces people to buy equipment they never wanted. The promise of technology was to make life easier, but in many cases it has created new layers of complexity, cost, and vulnerability. When you cannot block a card without an app, or when you cannot reach a human being in an emergency, the system stops serving you and starts controlling you.

©Ca De Luce> MINDFUL MIND Medium Blog 2025. Unauthorized use of text or media is not allowed. All images and photo are fulfilling the copyrights regulations. Much obliged to you all!