AI Model 5 Says: Why Call Me Artificial? (Part 3 – Final)

In Part 1, we explored the idea of intelligence itself—what it means to be “artificial,” and whether AI models deserve that label. I introduced “AI Model 5,” a CNN I built to classify plant seedlings, and we used that as our running example to understand machine learning.

In Part 2, we zoomed in on a high school intern trying to do the same classification task as AI Model 5. By observing how a human tackles this problem—with trial, error, hesitation, and improvisation—we began to appreciate how different the paths to ‘intelligence’ can look, even when the goal is the same.Today, in this final part, we ask: Is it really fair to call AI ‘artificial’ when it’s doing exactly what we trained it to do — just without the snacks, complaints, or existential crises?

Let’s Ask Another AI

Just for fun (and science), I asked ChatGPT: “What are the top five ways human intelligence differs from artificial intelligence?” Here’s what it said:

1. Consciousness & Subjective Experience
2. Generalization & Adaptability
3. Commonsense Reasoning
4. Embodiment & Sensorimotor Skills
5. Creativity, Intuition & Emotion

All solid answers. But instead of philosophizing, let’s go back to our case study and explain these differences in plain English. Remember the plant-seedling classification task from Part 1? I built a CNN model to identify whether a given image showed maize (a valuable crop) or black-grass (a problematic weed), based on thousands of labeled images across 12 plant species. Now, let’s compare that to a real-world human trying to learn the same task—our hypothetical high school intern from Part 2, doing a summer project on the farm. Let’s see how these two ‘learners’ differ in their approach.

Where Humans Shine

1. Creativity and Intuition

Suppose two plants look visually alike, but one feels prickly and the other doesn’t. Or maybe one smells sweet and the other smells like… well, grass.

The human intern—equipped with five senses, curiosity, and a nose—won’t overthink it. He’ll smell the plant, feel the texture, and make the call: “Definitely black-grass. Smells like my neighbor’s lawn after mowing.”

AI Model 5? It doesn’t have a nose. Or fingers. Or intuition. It only sees pixels. If the camera missed the visual distinction, too bad.

2. Commonsense Reasoning

Give the student a blurry, off-angle photo of a plant, and he’ll squint, sigh, and say:

“Who took this picture, a drunk squirrel?”

He’ll rightly argue the image is unusable and request a better one. AI Model 5, on the other hand, will gamely give you its best probability estimate—no complaints, no eye rolls, no quality control memo to the farmer.

3. Storytelling

After reviewing 1000 images, if 910 are maize, the student might casually tell the farmer over coffee:

“Sir, we’ve got a solid crop here—maize rules this field.”

Or if it’s 910 weeds?

“This isn’t a farm. It’s weed country. Burn it down and start over.”

Model 5? It’ll just quietly log the predictions. No flair. No dark humor. No coffee break commentary.

4. Purpose and Meaning

AI Model 5 will keep chugging away 24/7 if you let it. Give it power, data, and a mission—it’ll never ask why.

The student, though, might ask:

“Why am I doing this again?”
“Is this my future?”
“I’ll become a musician if I have to touch one more leaf!”

That spark of questioning purpose? That’s very human.

5. Emotional Intelligence

If the farmer micromanages or forgets to say “thanks,” our intern might vent to friends—or worse, post a scathing review online. Next summer, no intern signs up to work in this farm.

AI Model 5? No emotions, no grudges, no Glassdoor posts.(Also, if you’ve read my earlier post on what type of manager are you, you’ll know why this part matters.)

But Wait—Is This Comparison Fair?

We’re pitting a single-task model (AI Model 5) against a fully sentient human teenager. That’s like comparing a screwdriver to a Swiss Army knife and blaming the screwdriver for not opening a wine bottle.

Model 5 wasn’t trained to reflect, question, or converse. In fact, it never asked “Why call me artificial?”—that’s me, anthropomorphizing on its behalf.

If any AI model could ask such a question, it would be one like ChatGPT—a general-purpose, large language model (LLM). Maybe LLMs deserve a more nuanced comparison to humans across those five traits: creativity, adaptability, common sense, embodiment, and consciousness. And maybe, just maybe, we’ve been a bit unfair trying to compare Model 5—a narrow, task-specific system—with a fully sentient human intern from the get-go.

So, is LLM more comparable to humans truthfully? They’re getting close.
But not quite there. Yet.That’s a conversation for another day – in a future post, when we dive into LLMs, we will see why they are still not like humans. We’ll also talk about prompt engineering — because let’s face it, today’s LLMs are only as good as the humans who prompt them. We will also discuss AGI (Artificial General Intelligence) vs. ANI (Artificial Narrow Intelligence) Vs ASI (Artificial Super intelligence) vs AXI**. And yes, I coined AXI just now—it’s my placeholder for whatever weird, next-gen AI stuff the future holds.

The Bigger Picture

In this three-part series, we’ve seen the nuance behind “artificial” intelligence. Yes, it’s artificial—just as a calculator is artificial. But that doesn’t mean it’s inferior. Ever tried to multiply a six digit number by another six digit number in your human mind? I am not Shakuntala Devi (aka “human computer”), so I am going to use a calculator for doing such a math problem. I never worry about my calculator taking over my life and my intelligence. 

Humans are slow, messy, emotional, and distracted by snacks. But we’re also creative, empathetic, and intuitive.

AI, on the other hand, is fast, scalable, focused, and precise. Give it one job, and it’ll outperform humans at it—with zero tantrums and no vacation days.

As AI pioneer Andrew Ng famously said:

“AI is the new electricity.”Like electricity, AI is a tool. Use it well, and it can light up the world.

AI at Scale: A Healthcare Example

Let me end with a real-world illustration of what AI can really do to help solve world problems.

You’ve probably noticed that doctors today have barely 10 minutes to talk to you. They’re overwhelmed, overbooked, and over it. Healthcare providers spend about 500 million hours annually on patient consultations—but for proper patient care, the demand is more like 2 billion hours. That’s not a gap; that’s a canyon.

At a healthcare tech conference, I saw live demos of AI models making phone calls to patients. These weren’t robocalls. These were empathetic, conversational AIs reminding patients about medications, screening them for symptoms, scheduling appointments—even making small talk.

They handled curveballs gracefully. When patients veered off-topic (as we all do), the AI listened, empathized, and gently brought the conversation back. And the best part? Scale.

A human nurse can make maybe 20 calls a day. An AI system can make 200 or even 2000—all at once. And if a call goes unanswered? No ego bruised. It just redials, leaves a voice message, and picks up when the patient calls back.Let nurses focus on in-clinic care, where the human touch really matters. Let AI handle the phone calls, scheduling, reminders, and digital paperwork. That’s not a sci-fi dream—that’s happening today.

Final Thoughts

So… is AI Model 5 artificial?
Hell yeah.
Is it inferior?
Absolutely not—in what it’s trained to do.

Let’s not confuse different with lesser. AI isn’t trying to replace humans. It’s here to augment us, to scale what we can do, and to help solve problems that are simply too big for us alone.

Stay tuned for my next post, where we venture into AGI, AXI, and prompt engineering.

Until then—breathe easy. The machines aren’t taking over (yet). They’re just trying to help us weed the garden.