The part of artificial intelligence that nobody talks about in the breathless tech press or the Armageddon op-eds. It's no longer some distant science fiction cliché—it's right there in our pockets, finishing our sentences, tailoring our music, deciding what news we see, who we date, whether we get that loan. We're in the AI revolution whether we like it or not, and most of us are still pretending it's happening to someone else.
And we can't pretend anymore. We've been watching this happen for years now, and I have thoughts. Imperfect, conflicted, human thoughts on what we've made and where we're headed.
AI's promise from the beginning was freedom through automation. Free humans from drudgery so that we could focus on creativity, connection, meaning. Let the machines do the boring work—the data entry, the pattern recognition, the repetitive calculations—while we got on with the business of being beautifully, messily human.
I bought into that vision.
But somewhere between dream and deployment, something shifted. The tech we built with such good intentions got swept up in the churn of capitalism and power. The AI that was going to free us started to constrain us.
Walk through your day and count how many decisions are being made on your behalf by algorithms. I double-dare you.
Your alarm goes off at the time your sleep app calculated would optimize your REM cycles. Your phone serves up news articles chosen by AI to maximize your engagement. Your commute route is determined by traffic prediction algorithms. Your streaming service queues up content based on behavioral analysis of millions of users who clicked on similar things. Your social media feeds are coded by machine learning algorithms that know your preferences more than you know yourself.
At work, AI sorts through job applicants, monitors workers' productivity, flags potentially problematic emails. It decides which customers are eligible for credit, which patients are priority for doctors' appointments, which students are recommended for advanced courses. The AI never rests, never takes time off, never has bad days or personal biases.
Except, that is, that it certainly does have biases—they're just so deeply ingrained in the training data and algorithmic framework that we can barely discern them, let alone rectify them.
I've watched this happen in real time, and it's been like watching a car crash in slow motion. The AI systems we implemented were not inert tools waiting for human direction. They were active players reshaping the digital realm to their own logic, to their own optimization functions, to their own unknowable emergent behavior.
The untidy, raw, human internet I came of age with was replaced with something cleaner and more efficient. But efficiency is not always an upgrade. The AI web shows us what it thinks we want to see, creating echo chambers more sophisticated than anything human tribalism ever wrought. It optimizes for engagement, which it turns out is a euphemism for the fact that it optimizes for outrage, addiction, and senseless scrolling.
My nephew can get personalized calculus lessons at 3 AM, which is a miracle. But he's also being served a steady diet of algorithmically driven content designed to capture and sell his attention. The same system that's educating him is also learning him—mapping his interests, his weaknesses, his behavior patterns.
We created digital deities, and they're both more benevolent and more terrifying than we ever imagined.
The crazy twist? AI is consuming creativity whole, and we're cheering it on.
I've watched graphic designer friends lose clients to AI image generators. I've watched writers panic as chatbots produce content faster than any human ever could. Musicians are competing with AI composers that can mimic any style, produce infinite variations, never insist on royalties or creative feedback.
The response from the tech world has been as tone-deaf as you'd imagine: "Don't worry, AI will just take care of the boring things and allow humans to do the really creative stuff." But they're completely missing the point. The "boring stuff" is not somehow separate from creativity—it's part of it. The technical skill, the problem-solving, the happy accidents that happen when you're up to your elbows in the details of making something.
I learned to write by making terrible first drafts, wrestling with sentences that wouldn't obey, discovering my voice through thousands of hours of practice. Now I can generate a good essay outline in thirty seconds by prompting an AI. It's undeniably useful. It's also somehow hollow.
There is something in the struggle, the grit, the slow accumulation of skill that AI slices right through. When machines can produce technically proficient creative work in a moment, what happens to the human process of learning to be creative? What happens to the value we derive from creating things with our own hands, our own minds?
I am not glorifying struggle for the sake of struggle. I am simply worried that we are optimizing ourselves out of the very things that make us human. The virtual world is ever more a place where human creativity is made to feel obsolete, where the highest praise for human work is that it's "almost as good as AI."
Yet the creativity crisis is only the surface. Beneath it, AI has turned into the engine of the most advanced surveillance machinery in human history.
Every click, every pause, every micro-expression that your webcam sees gets piped into massive machine learning systems that build psychological profiles more detailed than those your best friends know. The AI not only knows what you bought—it knows when you hesitated before buying, what you looked at but didn't click on, how long you lingered over certain images or words.
This is not paranoid conjecture. I've helped build these systems. I've seen the data they collect, the conclusions they reach, the predictions they make regarding human behavior. The AI knows when you're going to break up with your girlfriend, when you're going to change jobs, when you're having issues with your mental health, when you're most vulnerable to certain types of manipulation.
And it's all commoditized. Your digital exhaust, your habits, your innermost fears and desires—they're all commoditized and sold to the highest bidder. The AI doesn't merely serve you advertisements; it shapes your reality, it dictates your decisions, manipulates your emotions in ways so subtle you don't even know it's happening.
The digital realm has become a war zone for the mind, and AI is the tool of choice. We are not merely users anymore but products and assets to be used for data and attention and behavior modification.
Meanwhile, AI is also quietly changing work in ways that are making the Industrial Revolution look quaint. This isn't just an issue of factory robots replacing assembly line workers. AI is coming for knowledge work, creative work, professional work that we thought was safely human.
I've witnessed whole customer service divisions being substituted with chatbots. Legal research that took armies of junior associates previously is now performed by AI in minutes. Medical diagnoses are more and more depending on machine learning programs that are able to identify patterns human physicians overlook.
The optimists say this will free us to do more valuable work, but they don't tell us what that valuable work is or who will pay for it. The pessimists predict massive unemployment and social collapse. My guess is the reality will be messier and slower—a slow erosion of human economic value disguised as technological progress.
What I worry about is that we're designing ourselves into irrelevance. Every task that AI can do better, faster, cheaper is a task that humans don't get paid to do anymore. Where do we get to the stage of being economic passengers in our own economy?
And then there's the issue of bias, quite possibly the cruelest irony of them all. We built AI to be a bit more objective than humans, to make decisions based on cold data rather than prejudice or emotion. Instead, we transferred all of our cumulative biases directly onto the silicon.
AI hiring algorithms are discriminatory against women and minorities because they're trained on historical hiring data that has decades of bias built in. Facial recognition algorithms identify white faces more correctly because the training sets were white. Criminal justice algorithms perpetuate racial disparities because they're trained on arrest and conviction data that reflects the bias of human judges and police officers.
The AI is biased against no one. It is not consciously biased. But it has learned to replicate human bias with mathematical precision, at scale, with the veil of objectivity that makes discrimination even harder to combat.
"The algorithm decided" is the new "we were just following orders." Bias is washed through mathematics and emerges on the other end as seemingly objective technical decisions. The online universe becomes less equal, not more, for all our yammering about AI democratizing opportunity.
Social media was exploiting our brains already before AI came to the party. But AI super-charged the manipulation to levels that feel almost extraterrestrial.
The content-recommendation algorithms know exactly what will get and keep you into a scroll, what news will infuriate you into engagement, what images will trigger your deepest desires or insecurities. They're not trying to entertain or enlighten you—they're trying to maximize engagement, and they've become appallingly good at it.
I've watched as loved ones were radicalized by YouTube algorithms that slowly, subtly nudged their recommended videos increasingly toward extremist content. I've watched as teenagers developed eating disorders from AI-curated Instagram feeds full of impossible beauty standards. I've watched as democracy itself creaked under the pressure of AI-amplified misinformation and polarization.
The digital space that was supposed to bring us together and democratize information has become an outrage, addiction, and division machine. And AI is the engine powering it all.
The thing is, though, I just cannot be a complete AI pessimist because I've also seen the actual miracles.
I've watched AI help paralyzed patients move robotic arms with their thoughts. I've watched machine learning algorithms detect cancer earlier and more accurately than physicians. I've watched AI translate languages in real time, breaking language barriers that existed for millennia.
In the initial period of the pandemic, AI assisted researchers in comprehending the virus, foreseeing its propagation, and creating vaccines in record time. Climatologists utilize AI to model intricate environmental systems and find solutions that we never could have otherwise. AI is assisting us in deciphering the human genome, finding new medications, and optimizing renewable energy systems.
The same technology that's hijacking our attention and automating away our jobs is also solving problems that can save millions of lives. The same algorithms that are propagating bias are also helping us to discover and correct for bias in human decision-making. The online world that AI is reshaping in unsettling ways is also being made more accessible, more personalized, more capable of meeting human needs.
This is what makes the AI question so infuriatingly complex. It is not a story of simple technological progress or dystopian collapse. It is both, simultaneously, and the tension between the two potentialities is tearing the very fabric of our culture apart.
They always wonder whether we must regulate AI, as if that's a simple policy question with a straightforward answer. But regulate what, exactly? The technology evolves faster than any government agency can comprehend, let alone control. By the time legislators get a handle on one generation of AI capability, the next one is already on the market.
And how do we control? AI is not a single entity—it's a massive ecosystem of techniques, applications, and implementations. Do we control the actual algorithms? The data they're trained on? The companies that implement them? The outputs they produce?
The real difficulty is economic and social regulation. How do we distribute the benefits of AI? How do we protect workers whose jobs are automated away? How do we prevent AI from concentrating too much power in the hands of a few tech giants? How do we maintain human agency in a world increasingly controlled by algorithms?
These are not questions for technology to answer. These are questions of what kind of society we want to live in.
Finally, my feelings about AI come down to my feelings about humanity itself.
I believe that human beings are creative, inquiring, compassionate beings who find meaning in striving, in connection, and in the slow process of becoming what we are meant to be. AI threatens to short-circuit or eliminate much of what it means to be human—the thrill of acquiring a skill, the intimacy of being completely understood by another sentient being, the dignity of productive work.
Yet I also believe human beings are resilient, resourceful, and capable of finding new sources of meaning when old ones collapse. Maybe we'll discover new forms of creating and connecting that we can't yet imagine. Maybe we'll learn to use AI as a tool for human thriving, not human substitution.
The virtual world is becoming something completely different in human history, one in which artificial intelligence defines reality at least as much as human intelligence. We are all in the beginning of this together, and none of us can predict how it ends.
So Where Do I Stand?
If you're going to make me pick a side on the AI discussion, I'm cautiously optimistic about the technology but extremely pessimistic about the way we're implementing it currently.
The problem is the economic and social systems that govern how AI gets used. We've built artificial intelligence in the likeness of capitalism: relentlessly optimizing, endlessly growing, seeing everything as an input to be exploited and monetized. Of course, it's consuming the world alive.
But it does not need to be so. We can design AI systems that prioritize human flourishing over profit maximization. We can develop algorithms that augment democracy instead of undermining it. We can use artificial intelligence to reduce inequality instead of amplifying it.
Would I stop AI if I had the chance? No. The potential benefits are too significant to lose. But would I fundamentally change the way we're developing and using AI? Without a doubt.
We need to start treating AI development like the civilization-altering force that it actually is. That means moving slowly enough to consider consequences before they become irreversible. That means involving everyone who's going to be affected by these systems, not just the entrepreneurs and engineers who are building them.
The digital space is becoming something new, something we've never experienced before. We're all figuring it out as we go, making it up as we build it. The stakes couldn't be higher, and the timeline couldn't be shorter.
But here's what makes me hopeful: for all its power, AI is still a human creation. We built it, we're controlling it, and we can still choose how to use it. The future is a choice we're making, every day, with every algorithm we install and every line of code we write.
We just need to start making that choice more consciously, more collectively, and with more attention to what we might lose in the process of gaining so much.
The electric gods we built are already among us. The question now is whether we'll remain their creators or become their creations. And that answer, thankfully, is still up to us.
