The most important dimensions of being human are the experiences we actually live.
AI systems require massive energy infrastructure while human brains operate on minimal power, yet still excel at emotional understanding, context, and social cognition machines cannot replicate. Art, literature, and human connection derive value from the imperfect, struggling creator behind them—not the technical output—making AI-generated content fundamentally hollow to audiences.
- Human brain operates on ~20 watts; training advanced AI requires hundreds of millions of watts
- AI-generated art proliferates online but audiences largely do not value it
- Beethoven composed his Ninth Symphony while almost completely deaf
- Research shows excessive interaction with computers and phones increases loneliness and sadness
A graduation speech argues that as AI excels at analytical tasks, distinctly human qualities—emotion, creativity, relationships, and lived experience—become more valuable, not less.
A speaker took the stage at a Washington graduation ceremony with a deliberate provocation. She announced she would not be talking about artificial intelligence—the two letters most likely to draw groans from a room full of young people in 2026. Instead, she said, she wanted to talk about human intelligence.
The framing was strategic. Every generation has watched transformative technology arrive with equal measures of awe and dread: the printing press, the steam engine, electricity, the internet. Each time, people asked the same question: what will be left for us to do? But that, the speaker suggested, might be the wrong question entirely. The better one is simpler: what does the existence of AI tell us about the things humans already do, and do in ways machines cannot replicate?
The answer, she argued, is deeply hopeful. Consider the basic physics of human cognition. A human brain weighs roughly three pounds and runs on about twenty watts of electricity—roughly the power needed to dimly light a refrigerator bulb. Training the world's most advanced AI systems requires data centers consuming hundreds of millions of watts, enough to power entire cities. These facilities sprawl across hundreds of acres, filled with massive servers, enormous cooling systems, and miles of cable. Meanwhile, a brain the size of a fist sits silently in a skull, using less energy than a laptop charger, doing things that still baffle machines. A small child can recognize a face in dim light, understand tone and emotion, navigate a crowded room, learn language through social interaction, infer intention, and grasp context—all without effort. Humans understand irony, ambiguity, affection, shame, love, guilt, humor, nostalgia. We can read the temperature of a room. We detect tension in silence. We sense insincerity behind a smile.
Machines excel at analysis. Humans do something else. We live in a complex world inhabited by other humans. The computer scientist Yann LeCun has pointed out that human intelligence is not simply computation. It is embodied experience, social understanding, and emotional cognition accumulated over millions of years of evolution. Perhaps, the speaker suggested, we should stop imagining humans as inferior computers. We have narrowed the definition of intelligence to precisely those forms of analytical reasoning that machines can optimize. But human consciousness is far richer and more mysterious than that. A machine can write a sad poem but cannot cry at a funeral. It can generate a love letter but cannot fall in love. It can describe fear but cannot lie awake at three in the morning wondering if it has wasted its life. And this matters because the most important dimensions of being human are the experiences we actually live.
The evidence is already visible online. Millions of AI-generated novels, essays, paintings, songs, and videos exist. Some are technically impressive. Most people do not care much about them. Why? Because art has as much to do with the human being behind it—perhaps more—as with the finished product. We connect with art because it comes from another human. When we read Charles Dickens or Toni Morrison or Gabriel García Márquez, we do not simply consume elegantly arranged words on a page. We enter their consciousness. We care that another human being struggled, suffered, imagined, doubted, hoped, and somehow transformed all of that into language. Those writings move us because they are, in a sense, imperfect. The Japanese aesthetic of wabi-sabi celebrates imperfection, incompleteness, asymmetry, transience, roughness, irregularity. Hand-thrown ceramics are prized precisely because they bear the mark of their maker: asymmetries, uneven glazes, deformations that reveal individuality and craft. Broken ceramics are sometimes repaired with gold seams in a practice called kintsugi. The crack shines instead of being hidden. The surface is rough, but what moves us is the visible evidence of fragility and repair.
For decades, society encouraged thinking of humans primarily as analytical machines. But perhaps AI is forcing a complete rethinking of that framework. If machines become far better than us at pure analysis, calculation, memorization, and pattern recognition, then what remains exclusively human becomes more visible, not less. In fact, the real danger of the AI era is not that machines become too human. It is that humans begin trying to become too much like machines. This is already happening. People increasingly talk about optimizing every dimension of life: sleep, productivity, networks, personal brand, performance. Students feel pressure to become perfectly designed résumés. Workers fear being measured against algorithms that never tire or sleep. But human flourishing has never consisted of optimization. Humans are imperfect, gloriously imperfect. A truly meaningful life is usually messy, nonlinear, contradictory, emotional, inefficient. The people who shape our lives most profoundly are rarely the most optimized. They are the most human: the teacher who inspired you because she genuinely cared about her students; the friend who sat with you for hours when your heart was broken; the mother who sacrificed decades for her child; the activist who refused to surrender even under threat of death; the scientist whose curiosity overcame failure.
Human greatness emerges from struggle. Perhaps one day a machine will write a technically perfect symphony. But it will never know the anguish of Ludwig van Beethoven, who composed his Ninth Symphony—one of the greatest musical works ever written—when he was almost completely deaf. When we listen to the Ninth, what moves us is not simply the arrangement of notes. It is the sorrow, perseverance, and triumph of a composer determined to create transcendent sounds he himself could never hear. This is why human intelligence matters. Not because it is faster than AI. Not because it is more efficient. But because it is rooted in lived experience. Machines can help us solve problems. But human intelligence is still what decides what is worth valuing, protecting, building, or sacrificing. The graduating class will live through extraordinary technological change. AI will transform medicine, science, education, transportation, communication, and perhaps every profession represented in that room. Some kinds of work will change. Entire economic sectors will evolve. But amid all this disruption, humans will still need what only other humans can offer. People will still want humans to educate their children, to help them overcome illness and pain, to console them in grief, to guide them in crisis, to create art about the human condition itself. And perhaps most importantly, people will still want to matter to each other. At its essence, human life is relational. We bond with other humans, for better and worse. We seek recognition, dignity, affection, and love from other humans. We cannot get any of that from computers or AI, no matter how powerful they become. The research is now overwhelming: too much interaction with computers and phones makes us lonelier and sadder. We have embraced technology and received the illusion of companionship without the demands or rewards of actual friendship. The speaker's hope for the graduates was simple: instead of competing with AI on its own terms, let AI push you to become more fully human.
Notable Quotes
The real danger of the AI era is not that machines become too human. It is that humans begin trying to become too much like machines.— Graduation speaker
We connect with art because it comes from another human. When we read Dickens or Morrison or García Márquez, we enter their consciousness and care that another human being struggled, suffered, imagined, and transformed all of that into language.— Graduation speaker
The Hearth Conversation Another angle on the story
Why does the speaker spend so much time on the physics of the brain—the twenty watts, the data centers? It seems like a tangent.
It's not a tangent at all. It's the foundation. She's saying: look at what humans do with almost nothing. We're not inferior computers that need to be faster or more efficient. We're something entirely different. The comparison makes that visible.
But couldn't you argue that AI will eventually do all those things—understand emotion, read a room—just with more computing power?
That's the trap she's warning against. She's saying those things aren't just harder versions of what machines do. They're categorically different. A machine might simulate understanding emotion. But it will never actually feel the weight of sitting with a grieving friend. That's not a limitation to overcome. That's the whole point.
She mentions wabi-sabi and kintsugi. Why does a broken ceramic bowl matter to this argument?
Because it proves that imperfection is not a bug we're trying to fix. It's the thing that makes something human and therefore valuable. A perfect bowl made by a machine is technically superior. But the cracked bowl with gold seams tells a story. It has been broken and repaired. That's what moves us.
The speech warns that humans are starting to optimize themselves like machines. Isn't that just the natural pressure of modern life? How do you resist it?
You resist it by remembering that the people who actually matter in your life—the ones who change you—are almost never the optimized ones. They're the ones who showed up messy, who struggled, who cared more about you than about their own efficiency. That's the model to follow, not the algorithm.
So the real argument is that AI doesn't threaten human value. It clarifies it.
Exactly. The more powerful AI becomes, the more obvious it is that what we actually need from each other has nothing to do with computation. We need to be seen. To be loved. To matter. Those things were always the point. AI just makes that impossible to deny.