The best teachers aren't optimized. They're remembered.
We're comfortable imagining AI replacing jobs. What we resist, almost instinctively, is imagining AI replacing the people who guide us. Politicians, teachers, and mentors. This issue is about why that resistance isn't just emotional. It might be one of the most important design constraints of our time.
AI can teach. But can it be a teacher?
When people talk about AI and jobs, the conversation usually follows a familiar path: Which jobs will AI replace? How many? How soon?
But there’s a curious blind spot in this conversation. We’re quick to imagine AI replacing jobs but hesitant to imagine AI replacing jobholders: people in leadership, influence, or authority.
We don’t talk about AI replacing politicians. We don’t seriously entertain the idea of AI presidents or AI mayors. Not because AI couldn’t, in theory, make better decisions, faster. But because something deep inside us resists the idea.
It’s not just about trust in AI. After all, politicians consistently rank among the least trusted professions. Members of Congress rank second to last in this 2024 Gallup survey, with only 8% of people rating them high or very high for their honesty and ethical standards.
Note that teachers rank second in this survey.
In fact, more people trust AI to go beyond providing information and make decisions or take actions, according to this YouGov survey from 2025.
The irony is rich: While we still distrust AI overall, we trust politicians less than machines, yet we would never vote for an algorithm. The question I want to explore today is… Why?
Because politics isn’t about logic. It’s about emotion
We don’t elect politicians for their policy platforms alone. We elect them for their ethos: who they are, or at least who they appear to be. Charisma, relatability, story. These are the real currencies of politics, not data and facts.
A leader’s power comes not just from competence but from connection. They represent us. They carry our values, emotions, and identities. An AI tool can optimize a transportation system, but can it speak at a funeral? Can it unite a country in grief or inspire it toward hope?
Science fiction has trained us to recoil at the idea. In Asimov’s The Evitable Conflict, a world governed by rational machines is technically perfect and quietly terrifying because humans have lost the ability to question it. Decades later, Black Mirror’s The Waldo Moment showed a digital cartoon character outcompeting human politicians simply by mocking them, a satire that feels less fictional every year.
The idea of being ruled by machines is almost always dystopian. We instinctively reject the idea of a society governed by rational intelligence alone.
But what about teachers?
What if we apply the same lens to education and wonder, with this in mind: What does replacing teachers with AI really mean to us?
Technically, AI can already teach lessons, answer questions, personalize feedback, and even detect when students are struggling. But just like with politicians, the answer isn’t about capability. It’s about what we value in the role.
Just like politics, education is not merely about delivering facts. It’s about shaping minds, guiding growth, and building trust. It’s emotional, relational, and deeply human.
Teachers are mentors, motivators, and role models. Students learn through them, not just from them. A teacher can sense when a student is having a rough day. Can lift someone’s confidence. Can build a classroom culture where learning feels safe and exciting.
AI can be a powerful tool in the classroom, but can it build relationships? Can it notice when a student is about to give up or light up when they finally get it?
If I learned one thing from my podcast Wide Walls, it’s that we remember teachers for their impact on us at an emotional level, not for the content they taught. When guests recall their favorite teachers, they share how they felt trusted, inspired, supported, and understood.
We know what the dystopian version of the future looks like. Rows of kids, eyes locked to screens, isolated from one another. Personalized content, zero connection. A sterile educational machine.
And yet, this dystopia creeps in. Not through evil, but through convenience.
So why keep humans in these roles?
Is this a lost battle? I don’t think so, and not just out of a simple moral conviction.
We keep humans in these roles because some decisions can’t be optimized. They have to be felt. Leadership and teaching require moral judgment, emotional nuance, and the kind of compassion that can’t be programmed. And we’ve been warned.
Our stories, from Asimov to Black Mirror, rarely imagine a world led by machines as anything other than cold and lonely. That instinct isn’t nostalgia. It’s telling us something real about what we actually need from the people who guide us.
Researcher and author Isabelle Hau makes this case beautifully in Love to Learn: the bonds between humans in the learning process aren’t peripheral to education; they’re central to it. If you want to hear Isabelle make the case herself, listen to the Wide Walls episode where she joined me to discuss her book.
This doesn’t mean AI has no place in politics or education. Quite the opposite. AI can be a partner. A fantastic assistant. A tool that makes human leaders and teachers more effective.
More importantly, maybe what we’re discovering isn’t that AI can’t replace these roles. It’s that we don’t want it to.
Not because of what AI lacks, but because of what it threatens to take away: the messy, emotional, human part of leadership, of learning, of being guided by another person.
Because the world is still composed of humans, and we, as humans, seek social connections. Certainly now, more than ever.
More to explore
- Love to Learn, The Transformative Power of Care and Connection in Early Education, Isabelle Hau
- Relational Intelligence, Educational Research, and Love to Learn with Isabelle Hau, Wide Walls Podcast
- The Waldo Moment, Black Mirror, S2
- The Evitable Conflict, Isaac Asimov
The article originally appeared on Substack.