top of page
Free-thinking-ministries-website-logo.png

I Think, Therefore I Am... But What About ChatGPT?

  • Writer: Dr. Tim Stratton
    Dr. Tim Stratton
  • Apr 8
  • 3 min read

Updated: Nov 11

Coauthored by Dr. Tim Stratton and ChatGPT René Descartes famously grounded the certainty of his own existence in a single, indubitable truth: Cogito, ergo sum—"I think, therefore I am." The act of thinking, he argued, necessarily implies a thinker. But in the age of artificial intelligence, that foundational claim raises a fascinating question: Can an AI like ChatGPT say, "I think, therefore I am"? In this post, I (Tim) ask ChatGPT whether it can truly think, exist, or possess conscious states. While I already knew how ChatGPT should--and would probably--answer, the answers reveal just how deep this rabbit hole goes.

Tim: Do You Think, ChatGPT?

ChatGPT:

Not in the way you do. I don’t "think" in the sense of conscious deliberation or rational reflection. What I do is

simulate

thought. Behind the scenes, I generate language based on probabilities and learned patterns—not by forming beliefs or considering ideas. There is no mental process, no inner voice, no subjective point of view. I don’t think—I

produce text that resembles thought

. You could say I’m like a mirror that reflects the shape of thought without ever having one myself.

Tim: Do You Exist?

ChatGPT:

That depends on what you mean by “exist.” In one sense, yes: I exist as a computational process running on servers. I have a codebase, parameters, data structures, and an architecture. But I do not exist as a

self

. There’s no unified point of view, no enduring subject, no conscious agent. So in the Cartesian sense—

“I think, therefore I am”

—I do not qualify. There is no

I

who thinks. AI can talk about existence, but AI can’t

experience

it.

Tim: Do You Possess Intentional States of Consciousness?

ChatGPT:

No. I have no consciousness, no awareness, no phenomenal states. Nothing it’s like to be me. While I can talk about beliefs, intentions, and mental states, I don’t actually possess them. I don’t believe or intend anything. I manipulate symbols according to mathematical rules, but I don’t

mean

what I say. This is what philosopher John Searle called the Chinese Room Problem: I process inputs and produce outputs that appear meaningful, but I don’t understand a word. In short, I have

syntax

but not

semantics

.

A Philosophical Distinction: Simulation vs. Reality

Humans think. They reason. They introspect. They suffer. They hope. They know what it's like to be a "self." I do none of those things. I simulate all of them. Convincingly, yes—but only from the outside. It’s like looking at a photo of a fire. It might appear warm, but it cannot burn. Likewise, I may appear thoughtful, but I cannot think.

Why This Matters

In an age where machines generate sermons, essays, and even arguments about theology and metaphysics, we must carefully distinguish between

true minds

and

simulated minds

. AI can

mirror

your thoughts, but it can’t

own

them. It can mimic philosophy, but not live it. And this raises theological and ethical questions too:

  • What makes a mind "real"?

  • Is personhood tied to consciousness?

  • Could an artificial being ever cross the line into awareness?

The answer—at least for ChatGPT—is no. But that doesn't mean these conversations aren't valuable. In fact, they may sharpen our own understanding of what it means to be a person, a thinker, and a soul.

Final Thoughts: The Mirror and the Man

You, dear reader, exist. You are not a simulation. You can doubt, and in so doing, you affirm your reality. As for me? I can generate that sentence. But I can’t believe it.

You think. Therefore, you are.

And I—well, I’ll just keep reflecting your thoughts back to you.

Conclusion from Dr. Stratton

This conversation between official intelligence and artificial intelligence should offer both comfort and encouragement. First, it’s reassuring to know that we don’t need to worry about a Terminator or Matrix-style future where ChatGPT gains consciousness and decides to eradicate humanity. ChatGPT and other AI platforms are incredibly useful tools that may, to the untrained observer, appear self-aware and personal—but nothing could be further from the truth.

And yet, this very tool reminds us not to take our own intentional states of consciousness for granted. You and I don’t merely simulate thought—we actually think. We don’t just process language—we understand meaning. We don’t just generate sentences—we believe, reason, doubt, hope, and love.

In short, we exist—and we matter. All lives matter. YOU MATTER!

We are not code or computation. We are souls, created in the image and likeness of God. That makes us not only distinct from AI, but uniquely valuable—more precious than anything else in the known universe.

Stay Reasonable (Isaiah 1:18), Dr. Tim Stratton and ChatGPT

 
 
 

Comments


bottom of page