Dreams and Artificial Stories

With the rise of AI, society must develop a deeper awareness of what is truly human—and what is merely a calculated probability.

Ever considered the difference between human dreams and the output produced by ChatGPT?

Let’s explore.

Machines That Calculate vs. Minds That Wander

ChatGPT is a generative pretrained transformer. It doesn’t think. Instead, it calculates probabilities: given a prompt, it predicts the most likely next word or phrase based on an enormous dataset of past human text. The result often appears fluid, grammatical, and relevant—sometimes even insightful. But it is not random. It is precise selection based on precedent and proximity within language.

Similarly, dreams also start from something—typically emotional residues or cognitive fragments from our day. Whether it’s anxiety about an exam, joy over a coming vacation, or the intellectual focus of coding a program, these experiences form the substrate of our dreams. And yet, what follows is vastly different.

Dreams develop. They evolve into creative, unpredictable narratives. They may reflect hope, fear, fantasy, or memory. They may defy logic, combine elements of time and space that never intersected, and invent things never seen before.

Generative AI can appear creative too. It can simulate style, invent stories, and generate poetic or even surreal responses—when prompted to do so. It can be instructed to act like a scientist, artist, or philosopher. It can weave a tale from the thinnest thread of a thought. But is that a dream?

What Thinking Really Means

Thinking is not just stringing together likely word patterns. It involves reasoning, abstraction, prioritization, reflection, sensation, and emotion. It is cognitive processing that includes both conscious and subconscious input—much of which comes from beyond the brain.

We think with our bodies too. The “gut feeling” is not metaphorical—it reflects real signals from our enteric nervous system. Emotions, memories, biochemical reactions, and physical states all influence cognition. Thinking, therefore, involves:

• Observing

• Evaluating

• Comparing

• Feeling

• Judging

• Imagining

• Mistaking

• Learning

And dreams are the playground where these faculties stretch and experiment. They are not just outputs—they are lived experiences of the inner world.

Why Machines Can’t Dream

ChatGPT does not make mistakes to learn from them. It does not feel. It does not observe itself. It does not care.

A machine might simulate the pattern of a dream. It might create a story that resembles something a human might invent. But it doesn’t dream—because it does not feel, reflect, or sense its own existence.

To truly “dream,” an AI would need:

• A nervous system or equivalent sensory feedback structure

• Emotional states that arise from its internal or external environment

• A capacity to reflect on its own relationship to its surroundings

• The ability to learn from failure, not just correction

• A sense of self, even if synthetic

AIs may soon produce outputs that mimic the form of thinking and dreaming. But for now, and likely for some time, they are sophisticated simulations—not human-like cognition.