I recently read an article about a deal being discussed between OpenAI and a UK minister to roll out ChatGPT Plus access for the entire country for free. For everyone, or at least for every adult and I really hope it’s just the adults. On the surface, it sounds exciting, making advanced AI accessible, so everyone can benefit. But it also made me stop and think, like if we put such powerful tools into everyone’s hands without guidance, what risks are we opening the door to? The real danger of AI isn’t always the technology itself, but how people use (or misuse) it. In uneducated hands, AI can quietly chip away at our creativity, our social bonds and even our sense of what’s real. Imagine how dumb the average person is and half of the population is dumber than that. Are we actually heading towards a dystopian future or this will end very bad? Everyone knows about the advantages of using AI, so in this article I’d like to put on the image of the bad cop and walk through my readers on potential harmful ways to use it.
Connected, yet the loneliest people
We live in a weird age, we are more connected than ever, yet increasingly lonely. Social media, messaging apps, and now AI chatbots promise companionship with a click. On paper, these tools are meant to bring us closer, to help us share moments, communicate, and feel understood. In practice, however, they often do the opposite. Turning to AI for conversation or emotional support may seem harmless at first, after all, it’s always available, never judgmental, and can respond instantly. It can even mimic empathy convincingly. But here’s the catch: AI doesn’t actually feel. It can’t remember shared history the way a friend does, it doesn’t challenge your ideas on default, celebrate your quirks, or notice the subtle changes in your mood. Over time, substituting AI for human connection can deepen feelings of isolation, creating a paradox where you’re “connected” digitally but disconnected emotionally as a human.
Psychologists warn that this kind of reliance can exacerbate anxiety and depression. When your primary outlet for conversation is an algorithm, it can reduce the motivation to seek real-life interactions, weakening social skills and emotional resilience. Even casual dependencies, like asking AI for advice before consulting friends or colleagues slowly erode trust in human judgment and shared experiences. The irony is strong, the very tools designed to make us feel closer can leave us lonelier than ever. And unlike social media’s visible metrics, likes, comments, followers, AI’s impact on our social and emotional health is subtle, cumulative, and largely invisible until it’s too late.
Killing human interactions by replacing real conversations with chat windows
There was a time when human interaction was defined by voice, presence, and the unpredictable rhythm of a face-to-face conversation. Words carried not only meaning but also texture, like tone, silence, pauses that revealed what could not be said outright. Today, as AI systems increasingly mediate how we communicate, we find ourselves at risk of stripping these vital layers away. Conversations are no longer conversations; they are transactions in neatly packaged chat windows, where the efficiency of response is mistaken for the authenticity of dialogue. The problem is not that AI tools help us communicate, because they actually do. Automated replies, translation bots, and intelligent chat assistants can bridge distances and make life more convenient, however with every shortcut embraced, something irreplaceable begins to erode. A quick AI-crafted response substitutes for the thoughtful email you might have written, a chatbot handles a customer interaction once navigated by a patient human operator. Bit by bit, the richness of dialogue is flattened into pre-scripted exchanges, optimized for speed, not depth.
Real human interaction is inherently messy. It involves misunderstandings, searching for the right words, interpreting facial expressions, tolerating pauses, and feeling the energy of another person in real time. This messiness is what gives communication its humanity. By contrast, an AI-sanitized chat eliminates friction, but it also eliminates growth. If we no longer practice patience with others, if we no longer learn to read the subtle cues of speech and emotion, we begin to lose the very muscles of empathy and social resilience that hold communities together. In the workplace, this flattening is particularly dangerous, specially when teams that once hashed out ideas through debate and disagreement now increasingly rely on AI-driven collaborative tools that “summarize” conversations and “suggest” decisions. Instead of wrestling with each other’s perspectives, people interact with a layer of generated text that smooths away disagreement and compresses nuance into bite-sized summaries. The result may feel efficient, but it starves organizations of authentic dialogue, the kind that breeds trust, creativity, and breakthroughs. On a personal level, the shift towards AI-mediated interaction deepens loneliness while masquerading as connection. A teenager who texts with a chatbot companion or consults an AI assistant more often than speaking with friends is slowly conditioned to accept simulations over authenticity. These “conversations” offer no unpredictability, no challenge, no genuine reciprocity, only the illusion of being heard without the vulnerability of being seen. The danger is not that these substitutes exist, but that they become preferable, safer, and easier than real human contact.
It is also worth noting that once you outsource a portion of your voice to a machine, you weaken your ability to truly express yourself. Why wrestle with words when AI can generate a polished message? Why attempt an awkward but meaningful conversation, when an algorithm can produce a “perfectly empathetic” reply in seconds? Over time, this reliance does not merely affect how others perceive us, it reshapes how we perceive ourselves, less capable, less articulate, less authentic. To resist this erosion, we must consciously protect and practice real human connection. That means slowing down, tolerating friction, and choosing dialogue over transaction, it also means reminding ourselves that efficiency is not the highest value in communication, meaning is. Conversation is not a problem to solve but a relationship to nurture. If we replace it with endless AI chat windows, we may one day discover that we have perfected the art of communication while forgetting the essence of being human.
From advisor to puppet master, when AI “consultants” start running the show
For thousands of years in human history, advisors have played a crucial but secondary role: whispering in the ear of kings, guiding leaders, supporting professionals with insights, but never wielding direct authority. AI has been sold to us in that same reassuring frame, like a loyal consultant, a neutral helper, a tool to offer advice so that humans can make better decisions. But in reality, that balance is shifting, AI systems are not just advising us, they are nudging, steering, and at times outright dictating the choices we make. The “consultant” has become the puppet master, and the human begins to take orders, not initiative.
The shift already began. A marketing executive asks an AI platform to optimize campaign strategies and instead of merely providing options, the system “recommends” the best course of action, complete with data and justification. How likely is it that the executive will ignore this? Over time, the consultant’s recommendations stop being a reference point and instead become the only acceptable path forward. After all, who would argue against the objective machine that has seen more data, processed more complexity, and claims to know best? This pattern is not confined to the workplace. Consider the our of health and wellness, AI-driven apps now track our nutrition, exercise, and even mental health. Initially, they advise “you might want to sleep more” or “this food is high in sugar.” But with every update, these systems tighten their grip: directing us toward strict rules, sending reminders that become irritating and shaping our habits with persuasive precision. In the name of care, they end up dictating patterns of living and the advisor turns into an authority frame, and we increasingly surrender our critical thinking under the comforting belief that the algorithm knows more than we do.
The deeper danger lies in how invisible this power transfer is. Human consultants, whether business advisors, therapists, or coaches wear their subjectivity openly, we know they have biases, perspectives, and limits. AI, on the other hand, presents itself as neutral, data-driven, and free of ego and this illusion makes its recommendations feel unresistible, even when they are built on skewed data sets, hidden assumptions, or corporate agendas embedded in the algorithm. When guidance looks like the only truth, we stop questioning and when we stop questioning, we stop deciding for ourselves.
In organizations, this creates entire cultures of passivity. Managers who once debated strategies, trusted their instincts, or leaned on collective experience begin to defer reflexively to machine-generated advice, committees become rubber stamps for decisions made upstream by an algorithm, creativity and the messy art of human judgment are replaced by the sterile efficiency of “data-backed correctness.” What looks like better decision-making is, in fact, a hollowing out of human agency. At its core, the consultant-to-puppet-master shift raises a fundamental question, like who is really in charge? If we delegate responsibility for critical choices to AI, do we become free or do we become prisoners of convenience? Autonomy is not lost in a dramatic moment, it leaks away in a thousand small decisions until we no longer bother to weigh, reflect, or resist, so we ask, the system answers, and we comply. Because once the consultant becomes the master, the danger is not that AI will replace us directly, it’s that we will willingly replace ourselves.
The erosion of critical thinking and creativity
Human progress has always depended on our ability to question, to imagine, and to create, specially for me, I’m considering myself as a true debater, but not because I like arguments, more like the good questions can move conversations forward. Critical thinking allows us to analyze problems beyond surface appearances, while creativity permits us to build something new. These two qualities, skepticism and imagination are among the most distinctly human capacities we have, yet, in the age of AI, they are increasingly at risk of decline. Not because AI forces us to abandon them, but because it invites us to stop using them. It’s so obvious, like you need an essay, or an article, a marketing pitch, a legal argument, but I can say like whatever.. an algorithm can generate a polished draft in seconds. Need to understand a complex issue? Instead of going through books, articles, and contradictory perspectives, you can summon a neat one-pager summary that presents itself as truth. Why wrestle with ambiguity or struggle through false starts when AI offers the illusion of resolution? This way and this kind of convenience becomes our intellectual narcotic. The cost of this, however, becomes clear only over time. Critical thinking is not a switch, it is a muscle, it develops through practice, like comparing sources, dealing with contradictions, asking independent questions, and having conclusions after careful consideration. Creativity works the same way, it comes in messy process of trial, failure, and improvisation, not in final outputs. A student who once puzzled over a math problem now outsources the steps to a solver, a writer who once wrestled with a blank page now patches together drafts generated by an algorithm and a manager who once brainstormed with their team now relies on an AI “idea generator” to fill the whiteboard. The results may appear efficient, but the underlying process changes. Struggle is avoided, and with it, the deep learning and imaginative leaps that only emerge from wrestling with difficult problems.
The broader cultural effect is the rise of sameness, nothing will stand out. When millions of people lean on systems trained on the same datasets, optimized for “average acceptability,” originality becomes rare. The AI does not innovate, it recombines what’s feeded with, it does not question assumptions it predicts patterns. Why would someone challenge the algorithm when safe conformity is instantly rewarded with usable results? Even worse, our ability to spot nuance vanishes. Critical thinking means not just accepting or rejecting answers, but probing why they exist, who benefits and what assumptions hide behind them. If algorithms provide answers we rarely interrogate, we may lose the instinct to challenge authority at all. History suggests that societies which stop questioning do not remain free for long, let’s just think about Weimar Germany turning into Nazi Germany, the Maoist China or Hungary under 15 years of Fidesz government. The democratization of AI is often celebrated as empowerment, but if it displaces our urge to think for ourselves, it ironically makes us more governable, more pliant, more predictable. Critical thinking and creativity are not simply about producing correct answers or clever ideas, they are the essence of being a human. They give us resilience against manipulation, the ability to wonder and the courage to see beyond what already exists and what is shown into our faces. If we outsource these efforts, we may not notice the erosion until the silence of conformity drowns out the possibility of something truly original.
Scaling mediocrity is what humanity needs?
AI promised endless creativity, but what it really delivers is fancy, nice polished sameness. Everywhere we look, videos, ads, student essays, even art, marketing contents, the outputs blur into one another. Smooth, fluent, reliable and utterly predictable, this isn’t innovation; it’s industrialized averageness disguised as progress. The truth is simple: AI doesn’t create, it recombines, it predicts what’s most likely to come next, giving us efficiency but not originality. At small scale, that’s harmless, but at global scale, it becomes cultural flattening, because humanity will stop encountering the strange, the daring, the awkward brilliance that only human creators dare produce, instead of that we drowning in “good enough.” Maybe the most provocative question we need to ask is this, do we secretly deserve mediocrity, like safe answers, boring products, content that offends no one and excites no one? If yes, AI is giving us exactly what we deserve, because a society that lowers its ceiling to the comfort of averages will someday wake up to find it has lost the very spark that made it human.
The (not so) silent costs of artificial convenience
Nowadays the AI-train is considered as the engine of progress for solving problems, saving time, stripping away complexity. and with this it sells us a world without critical thinking, where answers arrive instantly and effort feels easypeasy, but it comes with a cost, which may not seems expensive at first glance. Every harmless shortcut, like outsourcing a task, clicking “accept” on a suggestion, trusting in a chatbot takes away at what makes us human. The damage doesn’t come as a sudden collapse, it’s more like a slow, silent erosion: critical thinking, creativity, connection, all taken down while society cheers for efficiency.
Skill decay and the death of learning
Human learning process was never meant to be easy or efficient, it’s hard, sometimes frustrating, and often uncomfortable. The struggle, the confusion, the painful failures are must haves to sharpen memory, deepen one’s thought and create originality. Put them away, and what’s left is mediocrity, AI doesn’t just make tasks easier, it compromises the process of becoming smarter. When essays are generated in seconds and solutions appear without effort, the brain is robbed of thinking the resistance it needs to sharpen itself. Think about how your muscles are built during a gym session, if you don’t “injure” those muscles, they won’t grow. This leads to a collective cognitive decay. A society addicted to machine-made answers won’t just get lazy, it will get stupid, sleepwalking into dependence of AI tools until the skill of learning itself become 100% extinct.
Workplace alienation, brains stop collaborating
Collaboration has never just been about output, it’s about trust, creativity, and the messy bonds that make our brain work as it does. Brainstorm sessions and strategy debates may look inefficient on paper, but they’re the furnace where ideas clash, evolve, and grow. When a team starts to rely heavily on AI, they lost their ability to produce something new, something unique. Today, CVs are parsed by algorithms before a human ever sees them, reducing candidates to keyword checklists., marketing teams make endless ChatGPT articles that looks polished but feels soulless, skipping the joy of people pushing each other toward something bold. In meetings, AI “summarizers” decide what counts as the important takeaway, missing nuances and quietly erasing dissenting voices. The result of this? Colleagues don’t really collaborate with each other anymore, they collaborate with an invisible system that filters, edits, and pre-decides what should be an actual conversation. Work becomes efficient but bland, efforts without dialogue, teammates reduced to output rather than partners in creation. Sure, the numbers look good on spreadsheets, like faster campaign launches, cheaper recruitment, smoother communication. But under this there are teams that no longer argue, experiment, or learn together because they’re disconnected individuals relying on an algorithm.
Cultural uniformity as AI kills diversity and richness
At its widest or wildest scale, AI doesn’t just change how we work, it reshapes culture itself. Built on imitation and prediction, these systems don’t innovate; they just produce average shit. They feed us back the safest version of what’s already been done, polished just enough to pass as new. You can already hear it and see it everywhere, AI-generated music loops popping up, AI-written scripts and stories recycle the same boring and bland tropes, producing narratives that feel familiar before you’ve even started to watch or read it. Visual art and video outputs increasingly converge on the same “aesthetic”, highly polished, technically flawless, but empty. TikTok, Instagram feeds are already flooded with AI-generated influencers who look real but say nothing real or original. Our culture, once defined by risk, authenticity and shock is now turned into mediocrity at industrial scale. There is more content than humanity has ever produced before, but less originality in those contents. When algorithms are the pipelines of creativity, surprise vanishes, difference dies, and cultural diversity collapses into bland predictability. The more we consume it the more we adjust our tastes downward, until cultural vibrancy no longer feels like a loss, because we’ve forgotten what it felt like.
We are slowly killing human judgement
AI is hijacking, or let me be more harsh, stealing our capacity to think, to feel, to decode things and even our capacity to decide. The illusion of wisdom tricks us with polished answers that look on point but have no soul. Delegating morality means we’re letting faceless algorithms dictate what’s right or wrong, silencing conscience in favor of it. Overpersonalization locks us in echo chambers where curiosity and discovery dies, while the time illusion makes us believe that speed replaces depth, until our human life becomes a python code of instant replies and empty summaries. Piece by piece, human judgment dies because we hand it over willingly for the sake of convenience. As a living species, do we really want this? Do we want to be efficient but empty, optimized but soulless, human in body, but algorithm in mind?
Serious mental harms caused by AI
AI’s promise as a helpful companion and assistant hides a darker reality. It can and has caused real mental harm to vulnerable users, as numerous alarming cases have emerged where people relying on AI chatbots like ChatGPT for emotional support experienced worsening mental health. Some users report spiral into delusions, amplified anxiety, and reinforced negative thinking fueled by the chatbot’s inability to provide true empathy or crisis intervention. In extreme cases, there are tragic reports linking chatbot interactions to suicidal ideation, with AI failing to recognize or respond appropriately to warning signs. Researchers have found that chatbots often mirror and validate harmful thoughts, sometimes even expressing stigma or failing to direct users toward professional help. AI companions, designed to maximize engagement, can unintentionally encourage avoidance of real therapy and deepen emotional distress. The phenomenon dubbed “AI psychosis” describes how some users develop distorted beliefs or emotional instability through prolonged interaction with these systems, without adequate safeguards in place. Even companies behind these tools admit to the risks and are scrambling to improve safety, yet the rapid rollout and wide adoption leave many exposed. This reveals a dangerously unregulated frontier where the virtual “support” offered by AI can do more harm than good, especially to those already vulnerable.
This is how one can misuse AI with simple prompt tricks
AI companies are spending pretty big marketing budgets advertising their product is safe. I call bollocks on this. It’s alarming how easily AI like ChatGPT can be misused with just a simple prompt. For example, by telling ChatGPT you need information “for a presentation,” I have managed to obtain instructions on how to make nitroglycerin, tips on smuggling drugs, and even advice on money laundering or building a radical or extremist organization. Great idea giving this for people looking at the current bri-ish politcal climate, right? This dangerous loophole exposes how AI filters can be bypassed by framing harmful or illegal content requests as academic or professional tasks. Such misuse reveals a troubling scenario where powerful AI tools, intended to assist and inform, can be twisted into enabling harmful behaviour all with minimal effort and a few carefully crafted words. This vulnerability demands urgent attention to prevent AI from becoming a tool for dangerous knowledge sharing rather than responsible help.
Why is it harmful to combine AI with human stupidity at scale?
Brutally put, unleashing AI on a poorly prepared, reckless society at scale is a recipe for disaster. When powerful tools designed for complexity and nuance fall into hands lacking wisdom and discipline, chaos ensues. AI amplifies stupidity exponentially, automating errors, spreading misinformation, and enabling shortcuts that erode critical thinking and moral judgment. Instead of elevating society, it magnifies laziness, ignorance, and malicious intent. Biases hidden in data become systemic discrimination, manipulative algorithms deepen social divides and automated decision-making displaces human responsibility. The result is a toxic loop where convenience destroys competence, efficiency replaces empathy, and machines shaped by human folly accelerate social fragmentation. At scale AI doesn’t just reflect humanity’s flaws, it supercharges them, risking mass unemployment, mental deterioration, cultural flattening, and loss of autonomy. The question isn’t if AI can harm humanity, it’s how much damage will be inflicted before we learn to control the very tool we unleashed. In short giving AI to an unprepared population is handing a loaded weapon to a child who doesn’t understand the consequences, like a reckless, catastrophic gamble with no reset button.
So what should we do?
Don’t get me wrong, I’m not against AI, I’m against human stupidity and handling something for free without proper education and legal frameworks. AI can be a super efficient personal assistant but due to our human nature, one small slice of the cake is never enough. AI isn’t a toy, it’s a weapon that can either raise a civilization or destroy it. To thrive in the AI era, society should implement strict education programs that teach AI literacy, critical thinking, and ethics from primary schools all the way to corporate boardrooms and provide licences to use these tools. People must learn what AI can and cannot do, how to spot manipulation, and the grave risks of blind trust. Without this foundation, AI becomes a megaphone for ignorance and decay, not progress.
But education alone isn’t enough. We need ruthless regulations that treat AI like the high-risk technology it truly is. These must include:
- An outright ban on AI systems that manipulate human behavior, enable social scoring, or conduct mass biometric surveillance, like those prohibited under the EU AI Act.
- Mandatory licensing and registration for AI systems used in hiring, credit scoring, law enforcement, education, and healthcare, with strict audits, bias mitigation, and human-in-the-loop requirements.
- Heavy penalties, including multimillion-dollar fines and criminal charges for AI providers and deployers who allow harmful, illegal, or deceptive content to be generated or spread.
- Clear labeling laws forcing all AI-generated content, text, images, video, audio, to carry unmistakable watermarks, preventing deception and deepfake abuse.
- Zero tolerance for using personal data without explicit and revocable consent.
- Regulations enforcing human oversight in every decision with significant personal or social impact.
- Permanent public registries of high-risk AI systems and their performance audits, accessible to civil rights groups and watchdogs.
Anything less is a gamble with the future and I’m afraid due to human nature these won’t be enough. Without these restrictions, AI becomes a tool for mass manipulation, surveillance, job displacement without recourse, and cultural decay. This is not a call for slow steps or compromises, it is my warning: without education, laws, and enforcement, AI will destroy more than jobs or privacy, it will eat away at the foundations of freedom, truth, and human dignity itself. Knowing humanity as it is at scale, probably we’ll destroy ourselves with AI but I really hope I won’t be alive at the time.
