Common sense is the invisible architecture of human thought. It is a vast repository of practical knowledge that allows us to navigate a complex, physical world, a feat that remains beyond the reach of any system lacking an embodied existence.

Introduction

In the field of computational linguistics and cognitive science, researchers often encounter the 'Common Sense Knowledge Problem.' It is a paradox where Artificial Intelligence can perform high-level tasks, such as predicting market fluctuations or identifying microscopic anomalies in medical imaging, yet fails at tasks that a primary school student would find trivial. For instance, an AI may struggle to comprehend that if a physical object is placed inside a container, and that container is moved, the object moves with it.

To a machine, the world is a structured series of symbols, tokens, and statistical correlations. To a human, the world is an environment of weight, heat, resistance, and unwritten social expectations. We often take common sense for granted, perceiving it as a background process of the mind. In reality, it is a sophisticated, interconnected web of practical knowledge accumulated through millions of years of biological evolution and decades of personal, physical experience. AI lacks "skin in the game"; it exists in a vacuum of information, devoid of the embodied reality that anchors human reasoning.

Core Content

1. The Limitation of Symbolic Logic and Data-Centric Learning

For decades, the pursuit of Artificial General Intelligence (AGI) involved attempting to "program" common sense. Projects like Cyc spent over thirty years manually codifying millions of assertions about the world, such as "gravity pulls objects downward" or "you cannot be in two locations simultaneously."

  • The Infinite Scope of the Obvious: The primary obstacle to codifying common sense is its infinite nature. For every rule established, there are thousands of contextual exceptions that humans understand intuitively. We do not require a manual to know that a glass bottle will shatter on a stone floor but might survive on a carpet. AI requires explicit data or simulation to "know" this; humans understand it through the synthesis of physical intuition.
  • The Frame Problem: When assigned a task, an AI often struggles to determine what information is irrelevant. A human understands that if they are asked to "prepare a room for a guest," they should not remove the furniture to make the room "cleaner." Common sense acts as a critical filter, allowing humans to ignore the infinite irrelevant possibilities and focus on what is reasonable.
  • Intuitive Physics: From infancy, humans engage in constant physical experimentation. We develop a "feel" for the world, knowing how much force is required to push a door or the likely trajectory of a falling object. AI understands physics as a set of mathematical equations, but it lacks the intuitive grasp required to navigate an unstructured physical environment.

2. Embodied Cognition: Knowledge Through the Senses

The most significant divide between human and machine is embodiment. Human intelligence is not "detached" from the body; it is a product of it. This concept, known as embodied cognition, suggests that our thoughts are deeply shaped by our physical interactions.

  • Sensory Synthesis: An AI can provide a chemical analysis of a lemon and list its culinary applications, but it has no comprehension of "sourness." It has never experienced the physical reaction of the palate or the involuntary wince associated with the sensation. Without the sensory experience, the knowledge remains purely abstract and hollow. Real-world experience transforms data into genuine understanding.
  • Spatial Awareness: Humans possess a sophisticated mental map of the world. We understand the relationship between our bodies and the objects around us. We can navigate a dark room by relying on our memory of space and the tactile feedback of our surroundings. An AI relies on sensors and mapped coordinates, but it lacks the internalised sense of "being" in a space.
  • The Weight of Consequence: Human common sense is refined by the reality of consequences. If we ignore the laws of physics or social norms, we experience pain, hunger, or social ostracisation. AI operates in a consequence-free environment. It can provide a factually incorrect or dangerous response with total confidence because it faces no biological or physical penalty for its errors.

3. Social Common Sense: The Unspoken Protocols

Beyond the physical laws of the universe lies the intricate realm of social common sense, the unwritten protocols that govern human interaction.

  • Reading the Subtext: Common sense informs us when a conversation has become uncomfortable or when a "yes" actually implies hesitation. It is the ability to perceive the "social temperature" of a room. While AI can perform sentiment analysis on text, it cannot perceive the subtle energetic shifts, the lack of eye contact, or the heavy silence that defines a human crisis.
  • Practical Improvisation: In many professional settings, common sense manifests as the ability to find an improvised solution when standard resources are unavailable. A technician might use a temporary bypass to keep a system running during an emergency. This is practical knowledge born of necessity and experience. AI is rigid; it adheres to its programmed parameters. Human common sense is the guide when the parameters fail.
  • Cultural Shorthand: Every society possesses a common sense "shorthand"—a set of shared assumptions that do not need to be explained. This includes understanding the appropriate time for a joke or the unspoken expectations of a professional meeting. AI treats culture as a dataset to be queried, whereas humans treat it as a shared heartbeat.

4. Generalist Intelligence and Cross-Domain Transfer

Artificial Intelligence is typically a "specialist." A model trained for medical diagnosis cannot suddenly provide architectural advice. Humans, conversely, are "generalists."

  • Interdisciplinary Application: Our common sense allows us to take a principle learned in one area of life and apply it to an entirely different one. The patience learned in gardening may be applied to managing a complex team. This cross-domain transfer of wisdom is a hallmark of human intelligence.
  • The "Reasonableness" Test: In legal systems worldwide, the "Reasonable Person" standard is used to evaluate behaviour. It asks: "What would a person with ordinary common sense do in this situation?" AI cannot meet this standard because it lacks the concept of reasonableness. It only understands optimisation toward a defined goal, often at the expense of common sense.
  • Managing the "Edge Case": AI systems frequently fail when encountering "edge cases"—scenarios that were not present in their training data. For a human, life is a continuous series of unique scenarios. Common sense is the tool we use to navigate the unprecedented and the unexpected.

5. Why Life Experience Cannot Be Digitised

There is a prevailing belief that increasing the volume of data fed into a machine will eventually produce common sense. However, data is not a substitute for experience.

  • The Map and the Territory: Accessing a million descriptions of a storm is not equivalent to standing in one. The physical sensation of the wind, the scent of the rain, and the instinctive surge of adrenaline provide a form of "knowledge" that cannot be digitised.
  • The Role of Time and Maturation: Common sense is the result of years of filtration through a biological brain. It is the wisdom of an experienced professional who knows exactly how to handle a delicate situation without referring to a manual. This "feel" for the world is earned through time, not downloaded via an interface.
  • The Biological Anchor: Our fundamental needs, safety, belonging, and physical integrity, provide the framework for what is considered "sensible." A machine that does not require sleep, food, or safety will never truly understand the common sense of human priorities.

Conclusion

Common sense is the cohesive force that holds our reality together. It is perhaps the most sophisticated form of intelligence precisely because it appears so effortless to us. While Artificial Intelligence will continue to excel at solving structured, "hard" problems, it will continue to struggle with the "easy" ones, those that require a body, a history, and an investment in the physical world.

Your real-world experience is your most profound competitive advantage. Every challenge you have faced and every unwritten rule you have mastered has contributed to a superpower that no machine can replicate. You do not merely process the world; you understand it.

Artificial Intelligence is a processor in a void. You are a conscious being in a lived reality. Never undervalue the wisdom of your experience.

Quote

"Common sense is the knack of seeing things as they are, and doing things as they ought to be done."
— C.E. Stowe