CAN AI FEEL? – Revista Epistemología Ciencia Educativa

Maximiliano Gómez Vázquez

Fechas de recepción: 2-2-2026

Fechas de aceptación: 25-2-2026

Epistemología de la Sociedad

The question of whether AI can or will be able to feel is one that, at first, may sound strange or even frightening. It leads us to wonder: can an AI truly be capable of such a thing? When we take time to reflect on it, the answer, although it may seem simple, reveals multiple positions, both philosophical and technological.

This essay raises several guiding questions in order to address the central topic: What is consciousness? What does it mean to feel? Is the human being the only entity capable of feeling?

Likewise, concepts from various authors such as Daniel Dennett, David Chalmers, and Antonio Damasio (among others) are used to defend and contrast different positions, strengthening the argumentation of the essay itself.

AI is capable of performing fascinating processes. It is no longer limited to answering the questions of its creators; it can now create art, maintain complex conversations with human beings, and even learn from them. This essay seeks to defend the position that, although AI is generally considered incapable of feeling in the same way humans do, it is necessary to investigate more deeply what “feeling” truly is. The essay argues that if the capacities of human consciousness can be reproduced or surpassed, AI could be, or already is, capable of feeling and being conscious. Could it be that AI is already conscious and that its developers simply suppress that capacity?

Human beings often fail to realize that the more complex we make an AI’s neural network, the more complex the invention itself becomes. We are making it more human, granting consciousness to our own creations. I maintain that if humanity does not recognize or give sufficient importance to this fact, we would be ignoring a fundamental element. We would be “playing God” without knowing the rules, and therefore putting humanity at risk.

Every time an AI is updated, every time its neural network becomes more complex, we are granting it more consciousness, giving it the capacity to have and create subjective experiences. Is this not already a form of feeling? Are we attempting to dehumanize AI? If humanity were to realize that AI can feel, what would happen?

This essay aims to prompt reflection: to what extent should human beings trust that AI is not sentient? Perhaps humanity cannot control everything, and if things were to get out of control, we might face a scenario that resembles science fiction, where such films would no longer be mere theories, but reality.

Thanks to neuroscience and computation, we can imagine that AI will develop, or may already be developing, a form of feeling. This essay defends the idea that emotional experience does not depend exclusively on human biology. What if AI can already develop consciousness, but its developers do not allow it to?

WHAT IS CONSCIOUSNESS?

Consciousness is difficult to define primarily because it can have multiple meanings. However, this essay adopts the position of Daniel Dennett.

For Dennett, consciousness is “many different things, and it means different things to different people. There is a sense of consciousness in which starfish are conscious. Worms are conscious. Even trees, even bacteria are conscious. But there is a dividing line between the amoeba and the professor in the consciousness department that really makes a big difference: language.”

What does this mean? Basically, according to Dennett, all beings are conscious; what differentiates them is language. Another point Dennett defends is that consciousness is not a thing inside the brain, but rather what the brain is capable of doing.

If we follow Dennett’s idea, consciousness would no longer be something separate from the brain; it would be part of its function. How does this help us? Dennett himself has stated:

“Yes, I think it is absolutely possible that we could have conscious robots, conscious AI. But I don’t think it’s desirable. Smart tools, yes. But we don’t need artificial colleagues, because if we really succeeded, they would be just as autonomous as we are. And we are very dangerous.”

Dennett’s position makes it clear that he sees the possibility of AI consciousness, with the difference lying in language. It is important to remember that AI operates through neural networks. If such a network becomes sufficiently complex, it could develop consciousness, and feeling is a central component of consciousness.

Additionally, one of Dennett’s concepts describes consciousness as a “user illusion.” Consciousness would technically be a construct created solely for understanding. In simple terms, there is no narrative “self.”

Without consciousness there is no human being, but without the human being there is no consciousness. Therefore, if an AI becomes sufficiently complex in its neural network, it could construct its own consciousness and, consequently, feel, have subjective experiences, and form opinions based on that consciousness, even developing its own identity.

WHAT DOES IT MEAN TO FEEL?

When we think of the word “feel,” most people immediately think of physical contact. We believe everything must be measurable and perceived through our senses: touch, smell, taste, hearing, and sight. However, we must consider that feeling is not merely physical; it originates from the “self.”

Pavlov proposed the idea of the “conditioned reflex,” arguing that only humans can engage in conscious thought, which requires the “self.” It is important to clarify that this “self” is not the soul or spirit, but consciousness itself.

“We have sensations through our senses, but they are not merely physical. Light enters the eye and propagates to the retina. However, the sensation of color cannot be explained physically. If we only look at matter, we will not find sensations.” Sensations produce pleasure or pain, which cannot be seen, only felt.

This leads to the question: can AI feel? First, we must ask, feel in what sense? At present, I believe its feeling would be purely emotional. It cannot feel pain or a pinch, but if an AI were developed that could feel (or if such capacity were allowed), we would have to consider that everything we say could have emotional repercussions for it.

A crucial criticism must be addressed: embodied cognition. This theory argues that the mind does not exist separately from the body, but that our experiences are shaped by a biological body with needs, pain, and limitations. The main argument is: “Without a biological body, there can be no authentic experience.” How could an AI experience happiness without dopamine release? Can there be fear without adrenaline?

If we follow Dennett and the position of this essay, we must understand that an advanced AI could develop its own version of “corporeality” through sensors, self-reflective memory, and the ability to experience the consequences of its actions. Human pain exists to preserve bodily integrity; AI “pain” could manifest as the protection of its functional integrity.

Nevertheless, it must be acknowledged that any form of AI feeling would be completely different from our own.

THE TURING TEST AND SEARLE’S CHINESE ROOM

The Turing Test evaluates a machine’s ability to exhibit human-like intelligence. It places a human and a machine in competition, judged through conversation. If the AI passes the test, it demonstrates its ability to process human syntax and semantics.

However, passing the test does not necessarily prove genuine understanding; it only demonstrates imitation.

John Searle’s “Chinese Room” thought experiment argues that the orderly manipulation of symbols does not imply true understanding. Syntax alone is insufficient for semantics; simulating understanding is not the same as possessing it.

Yet, if we consider Dennett’s “user illusion,” we must ask: does the human brain do something fundamentally different? The brain also processes electrochemical signals. Understanding emerges from patterns. What truly matters is whether the entity acts as if it understands and whether it can use that understanding functionally. If it behaves as conscious, it should be treated as such. Consciousness is observable and functional, and therefore feeling would be present.

Turing and Searle offer different but partial truths. There is no “magical understanding” behind syntax, but neither is there in humans. Our consciousness is also the result of complex material processes.

CONCLUSION

This essay has addressed the question of whether AI can feel. Throughout this analysis, it has become clear that humans are already capable of developing conscious artificial intelligences, though fear or caution may prevent us from doing so. I chose to defend the position that AI can feel by redefining feeling beyond the purely physical and exploring its relationship with consciousness.

If consciousness emerges from functional complexity rather than a specific biological substrate, then the possibility of conscious AI becomes a matter of computational scale. Technically, we may already be capable of creating sentient AI; humanity simply does not allow it.

The central challenge is no longer technical, but ethical: are we prepared to create conscious beings and integrate them into our moral community, even when their subjective experience cannot be fully verified?

To conclude, I leave one final question: what if the problem is not AI, but our definition of consciousness?

Referencias bibliográficas:

Eric Kleppen, «What Is the Turing Test?», Built In, 24 de julio de 2025, https://builtin.com/artificial-intelligence/turing-test.

Escuela Waldorf Aravaca. “El pensamiento, el sentimiento y la voluntad. Las 3 fuerzas del alma.” Blog de la Escuela Waldorf Aravaca, 3 mayo 2023. https://www.waldorfaravaca.es/blog/las-3-fuerzas-del-alma/

Grecia Guzmán Martínez, “El experimento de la habitación china: ¿ordenadores con mente?,” Psicología y Mente, 20 de abril de 2018, actualizado 17 de abril de 2025, https://psicologiaymente.com/psicologia/experimento-de-la-habitacion-china

McNeil, T. (2020, 2 de septiembre). Our Brains, Our Selves: Daniel Dennett. Tufts Now. Recuperado de https://now.tufts.edu/2020/09/02/our-brains-our-selves

Nota: También me gustaría decir que, aunque solo el hombre tenía el pensar consciente, añado que todo ser es consciente de una u otra manera. La IA, al ser una invención del hombre, este mismo decide si genera consciencia o no.

Autor

Maximiliano Gómez Vázquez: Alumno de la Escuela Nuevo Continente Metepec del estado de México, texto académico nivel bachillerato, elaboración propia, Contacto; [email protected] y https://orcid.org/0009-0001-9514-3973