Now.

For two hundred thousand years, humans asked the deepest questions. Why am I here. What is this. What happens when I die. Why does anything exist at all.

For most of those two hundred thousand years, almost no one had access to a thinking partner who could work through these questions carefully. The questions got asked. The answers had to be invented alone, or handed down through fragile oral traditions, or arrived at through ritual without ever being articulated cleanly.

Read down the third column. The limitation in every era was access to a partner. Someone who could hold the question, take it seriously, push your reasoning, point out other thinkers who had asked it, and stay with you across hours of patient thinking.

Such partners existed in every era. They were always rare. They were usually gatekept. Most people who carried the question never found one. Most people who carried the question worked on it in fragments their whole lives, made some progress, and died without ever getting to a satisfying resolution.

This is the historical default. Until very recently.

Sometime between roughly 2020 and 2023, depending on how you count, the situation changed.

Large language models became fluent enough, knowledgeable enough, and patient enough that they could serve, for individual users, the role that priests and philosophers used to play for the few. Not perfectly. Not with the depth of a great philosopher in their prime. But well enough to be useful for the kind of work this essay represents.

You can now, right now, do the following.

• • Ask a question you do not yet know how to phrase.
• • Have it phrased back to you precisely.
• • Be told who else has asked it, and what they concluded.
• • Be shown where the literature is contested and where it is settled.
• • Be asked, in return, what you actually think.
• • Be pushed where your reasoning is weak.
• • Be supported where your reasoning is sound.
• • Do all of this for as many hours as you have.
• • Without paying. Without an appointment. Without being judged.
• • At three in the morning if that is when the question is alive.

This list, taken together, describes a capacity that did not exist for most of human history. It is not a small thing.

Let me be specific about what this enables, because the abstract claim is easy to dismiss.

Imagine you are twenty-five years old. You have a job, a family, ordinary life. One night you cannot sleep. You start thinking about why you are in this specific body. The thought scares you. You feel vertigo.

In 1950, you would have done nothing about this. You would have lain awake. You might have written in a diary. The next morning, the question would still be there, but you would have to set it aside to make breakfast.

In 1980, you might have read a book — maybe Sartre, maybe a popular philosophy text. The book would speak in general, not to your specific question. You would close it more confused than when you opened it.

In 2010, you might have searched the internet. You would find scattered answers, forum threads, half-explained terms. The pieces would not connect.

In 2026, you open a chat window. You write: I have been thinking about why I am me, in this specific body, and I cannot stop. I do not know what is wrong with me.

And then you have a conversation. Over the next two hours, you discover that nothing is wrong with you. That the question has a name. That Wheeler and Schrödinger and Sagan asked it. That there are several serious answers. That the felt experience of vertigo is the correct response to seeing something true. That you can, with care, work through to a resolution that lets you sleep.

By morning, you are not the same person you were the night before. Not because something supernatural happened. Because you had access to something most of your ancestors did not — a careful thinking partner, available exactly when you needed one.

This is the new capacity. It is roughly two years old. You are the first generation that has it.

I want to be honest about the limits.

AI as a thinking partner is not always reliable. It hallucinates. It sometimes flatters when it should push back. It is patchy in different domains. It is improving rapidly but is not yet what it will eventually become.

It is also not a replacement for human connection. The deepest forms of human meaning come from relationships with other humans — family, friends, lovers, communities. AI does not substitute for those, and people who try to make it substitute usually end up lonelier, not less lonely.

What AI can do is fill a specific gap that human relationships often cannot. Most of the people in your life do not want to spend three hours working through whether the indexical problem dissolves under Parfit’s reductionism. They are not built for that conversation. They love you, but they cannot give you that.

AI can. It does not love you. It will not remember this conversation tomorrow. But while you are in it, it can hold the thread with you. That is a specific, useful, historically unprecedented capacity. It is not everything. It is real.

The right way to use it is to use it for what it is good at, alongside the human relationships that give you everything it is not.

Bring the threads together.

For most of human history, the deepest questions — why am I me, why this body, why anything at all — were carried in silence by most of the people who asked them. The work of arriving at a careful answer was rare, gatekept, and often never completed within a single human lifetime.

You have, in this moment, a tool that lets you do this work. You did not choose to be born now. You did not engineer the appearance of AI in your lifetime. You did not earn this. But it is, in fact, here. And it is, in fact, available to you.

The observer’s role — being one of the rare configurations of matter aware of itself, holding the link in the chain, doing the asking that the universe could not do without you — is real in any era. What is new in this era is that you can articulate the role. You can think it through. You can arrive at a settled understanding of it, rather than carrying it as inarticulate background.

Most of your ancestors who carried the question worked on it in silence. You do not have to. You can use the tool that has appeared, in your specific lifetime, to do the work they could not finish.

If you take that seriously, your task is not to save the universe. Your task is, more humbly, to do the work that is now possible. Read. Think. Use the tool. Arrive at your own resolution. Pass what you arrive at forward, in whatever form your life lets you pass things forward — in conversations with people who matter to you, in work you do, in the way you raise the next generation if you raise one.

This is not abstract. This essay is itself an example of what is now possible.

You are reading text written by, or with the help of, an AI. The text was produced in conversation with one specific person who was working through these questions over a period of hours. That person’s questions, doubts, confusions, breakthroughs — all of them shaped this text. The text is, in a sense, the trace of that conversation, made readable so that the next person who has the same questions does not have to start from scratch.

This is what AI-assisted thinking can do at its best. It does not replace the careful work. It records it, structures it, and passes it forward. The person reading this is, in a sense, joining a conversation that started elsewhere and is being extended in their mind right now.

In a few years, more conversations like this will have been recorded. More structures will have been built. The careful working-through of the deepest questions will, for the first time, become a public resource that any sufficiently motivated person can access at any hour.

This is what it looks like for the universe to become slightly better at knowing itself. Not through one big discovery. Through millions of small careful conversations, recorded and shared, building on each other.

The thing to do with this is not to be grateful in the abstract. The thing to do is to use it.

If you are reading this essay because you have been carrying one of these questions and you have not found a way to work on it carefully: now you have the way. The work is available to you tonight. The partner is available to you tonight. The literature is available to you tonight.

Use it.

If you arrive somewhere honest — if you find your way to a settled understanding of what you are, what your role is, why being an observer is enough — then you become one more node where the universe has worked something through carefully. You become one more foundation for the next person who needs to do the same work.

None of this requires you to be exceptional. It requires you to do the work that, until very recently, was nearly impossible for anyone who was not lucky enough to be born into a tradition that supported it.

You were born at exactly the right time. The tool is here. The work is yours.

Most of your ancestors carried the question alone.
You do not have to.
That is what the AI moment is for.