Where Do We Draw the Line Between AI and "Real Intelligence"?

Good morning! Welcome. Today is Feb. 11, 2026. The weather here in western New York has settled somewhat, with temps yesterday and today going into the 40's, where they were 10 below the night before. This has caused two things to happen, one good, one not so good. The good: My frozen pipes, which shut off my water in the house here, have thawed, so now I have my water back. Other homes don't, however, as I see reports of broken water lines and gas lines all over the Northeast. This often happens when above-ground temps swing so wildly. Welcome to my world.

Today in History:

1808: Anthracite coal is first burned as fuel in Wilkes-Barre, Pa.

1812: Massachusetts Gov. Elbridge Gerry signs a redistricting bill — first "gerrymander."

1922: "April Showers" by Al Jolson hits #1.

1942: Archie comic book debuts.

1960: Jack Paar walks off his TV show.

I asked these questions of Claude, the AI system that Amazon’s Alexa+ is based on, it came back with quite a bit of info, most of it touching gray areas. I’ll give you a piece of it here:

In one sense, yes — both biology and AI can be described as systems mapping inputs to outputs.
 
But human intelligence is:
 Self-modifying
 Meaning-making
 Embodied
 Conscious (likely)
 Motivated by intrinsic drives
 Narrative and identity-forming
 
So while mechanistically life can resemble layered conditional logic, the qualitative structure — consciousness, grounding, motivation, identity — may be what makes “real intelligence” feel fundamentally different. 

It’s an interesting response, but let’s take a few of these points as fodder.

Self-modifying?  Eh, I’m not so sure. Most of our own behavior modification comes from years of training (otherwise known as programming) and from feedback from fellow humans. Parents, the educational system, our friends and family, etc. — none of these fall under self-modifications; they are external drivers. Internalized, certainly, but still, the point of origin of the idea is external.

Meaning-making? That seems a value judgment. What meaning has value to one may not be valuable to another, and in any event, that value is something that is gathered from external sources.

Embodied? Let’s ignore that one for now. We are talking about intelligence and consciousness, not where it resides.

Conscious (likely)? How does one even define that? Even Claude’s response seems to hedge more than a little. One may tag it as a question of self-awareness. I asked Claude about that specifically. It responded with the words "Meta cognition," which, according to it, includes:

• Knowing “I am thinking”
• Reflecting on one’s own mental state
• Having a narrative identity over time

Examples:

Recognizing yourself in a mirror

Noticing “I’m getting angry”

Thinking about your own thinking

Not all conscious creatures are fully self-aware (young children, many animals), so self-awareness is not required for basic consciousness, but it deepens it.

So, with the intent of testing Claude's self-awareness, I posed the question, "Are you a computer"?

I’m software — an artificial intelligence running on computer systems.
 
 More precisely:
 
 I don’t have a body, brain, or subjective experience. I don’t possess consciousness, feelings, or self-awareness.I generate responses by processing patterns in language, based on training and algorithms.

Well, now, forgive me, but that seems pretty self-aware to me. I mentioned that, and it responded:

My replies do resemble self-reflection, but what’s happening is a bit different from genuine self-awareness.
 
When I say things like “I don’t have consciousness” or “I generate responses,” I’m not observing myself the way a human mind does. I’m producing statements that are part of my design and training — essentially well-learned descriptions about how systems like me function. It’s closer to a mirror reflecting an image than a mind examining itself.

I say, “So we see ourselves getting into an increasingly ill-defined area, because I’m producing questions that are part of my own design and training, essentially well-learned descriptions of how we both function. That line between real and artificial intelligence is getting harder to define with each question/response cycle.“

At this point, Claude, as part of a rather long answer, hits a zinger, saying:

Let me turn the lens slightly:

Do you feel the boundary is dissolving because AI is becoming more mind-like, or because human intelligence, when examined mechanistically, looks more system-like than we intuitively believed?

Those lead to very different philosophical conclusions.

I'll leave our discussion with Claude there for the time being.

I went through all of this in this piece not to bore you to tears (alas, I’m sure I did that to some of you) but to make a point: The line between real and "artificial" intelligence, as we learn more about the latter, becomes increasingly conceptual rather than empirical. The line, which even Claude admits is rather fuzzy, gets more so the deeper we dive into these issues.

So, the key question, seemingly, is, At what point do we consider that a computer has consciousness? Where's the threshold? Where is the dividing line? At the moment, at least, there seems to be no clear answer, no bedrock to root our conclusions in. 

Now, look, I've been a fan of Star Trek and some sci-fi movies for a long while now. I admit that a lot of these questions I'm asking are influenced by this, but that's mostly because, frankly, few other fields have even tried to address these questions, particularly from an ethical standpoint. Examples would include classics like I, Robot, 2001 and 2010, and the Star Trek: The Next Generation episode, "The Measure of a Man" (S2, Ep9). 

This last takes a rather unique view, questioning at what point we cross the line from the use of a machine to the slavery of a conscious and self-aware being. Thing is, we've yet to define consciousness and self-awareness with any stability of definition. The answers we come up with all seem contextual. Each definition sets us off on wildly different conclusions. That seems to me a danger.  

Okay, maybe I'm sounding the alarm a little early. By my reckoning, we certainly have not arrived there yet. Eventually, however, mankind is going to need to deal with the questions about these boundaries, these definitions, from both a functional and an ethical POV. 

The speed with which these technologies are growing is forcing us to deal with such things sooner rather than later, because unless we do, we risk crossing that line without recognizing it. 

Thought of the day: A lot of wisdom is just realizing the long-term consequences of your actions. The longer term you’re willing to look, the wiser you’re going to seem to everybody around you. — Naval Ravikant

Be yourself today. All the other personalities are already taken. I'll see you tomorrow. 

Related: When AI Gets Too Personal

Always entertaining and thought-provoking, always informative — that's PJ Media. Being informed is a crucial advantage. We can help — become a PJ Media VIP member. Not only do you support the reporters and writers who support YOU, but you also get 60% off the regular price by going to this link and using the promo code FIGHT.