Psychology of Singularity – Your Source for AI Information

Image by Hotpot.ai

Beyond Word Prediction: The Limits of Artificial Intelligence and the Superiority of Human Cognition

By Jill Maschio, PhD

March 10, 2026

We have probably heard comments on the Internet claiming that AI is more intelligent than humans. What needs to be clearly stated is that AI may be more intelligent than humans in some areas; otherwise, this is a misconception. I propose that the claim that AI is overall more intelligent than humans is a myth and a form of hype. Let me start by paraphrasing Noah Yuval Harari at Davos in 2026. Harari, a historian and renowned author on human evolution and artificial intelligence, was a guest speaker at Davos. His speech painted a picture of the future in which AI is an autonomous agent that can make decisions, predict the next word in order, and do so far better than most humans can. Speaking a sophisticated language is one of humanity’s defining strengths, enabling us to express our thoughts. Words are more than letters on a piece of paper. If I have learned one thing in life, it is not the words that matter the most; it is how we interpret and perceive them. How we perceive it will influence how we choose to respond, making our behavioral responses equal, if not more, important. Someone can tell you that they love you, but it is the endless caring and support that gives it true meaning – that is what makes the relationship grow and flourish.

Harari further suggested that artificial intelligence could function like a knife—capable of cutting a salad or committing murder depending on how it is used. While the analogy highlights the dual-use nature of technology, it is also problematic because AI systems lack moral reasoning. Any ethical constraints embedded in AI originate from the programmers, designers, and data that shape its algorithms. In a public interview, when commentator Tucker Carlson asked Sam Altman how ChatGPT makes moral decisions, Altman suggested that aspects of his own experiences influence the system’s design. This perspective raises important concerns about the degree of authority society may grant to artificial systems whose outputs are ultimately rooted in human-coded rules and training data rather than independent understanding. If humans increasingly defer judgment and decision-making to AI, there is a risk of diminishing the very cognitive capacities that define human intelligence. The central argument of this article is that human cognition—shaped by consciousness, meaning-making, lived experience, and moral reflection—remains fundamentally distinct from and more comprehensive than the algorithmic processes underlying artificial intelligence. The following sections outline ten reasons why human intelligence continues to surpass AI.

The debate over artificial intelligence often centers on its impressive computational abilities, particularly its capacity to generate fluent language and to process, sift through, or analyze enormous quantities of information. However, equating these capabilities with genuine intelligence overlooks several fundamental aspects of human cognition. Human intelligence is not limited to producing coherent sequences of words or identifying statistical patterns in data. Rather, it involves consciousness, meaning-making, emotional understanding, moral reasoning, creativity, and the integration of lived experience. When these broader dimensions of cognition are considered, the differences between human intelligence and algorithmic systems become much clearer. The following ten points outline key reasons why human cognition remains fundamentally more sophisticated than current forms of artificial intelligence.

One

First, I argue that the intertwining of human cognition, thoughts, and language is a central feature of human uniqueness. We should value what makes the human mind unique, which is partly how the brain functions both with and without our full awareness. Sigmund Freud conceptualized that there are two minds that operate in parallel. One is the conscious mind, which allows us to hear and know what we are thinking, and the other is the unconscious mind, which operates automatically and independently. The two minds cooperate to accomplish a goal, with the unconscious exerting influence over the conscious without the person realizing it (Custers & Henk, 2010). The unconscious mind illustrates a fundamental difference between humans and AI, making humans superior in this context.

Harari emphasizes that AI can produce a string of words faster than humans can. Research from the California Institute of Technology (Zheng & Meister, 2025) indicates that humans process and act on conscious thoughts at approximately 10 bits per second. Humans may perceive approximately 1 billion bits per second, and the amygdala can detect potential threats in as little as 20 to 100 milliseconds – faster than the blink of an eye. The ultra-fast subcortical pathway allows the brain to prioritize survival states before the person is consciously aware of a potential threat (Wang et al., 2023).

In a study by Whalen et al. (1998), the amygdala was shown to detect emotional facial expressions without a person’s explicit knowledge. In their experiment, the researchers used functional magnetic resonance imaging (fMRI) of the human brain while showing participants pictures of faces with fearful or happy expressions, using a “backward masking” technique. The emotional faces were displayed for just 33 milliseconds, immediately followed by a neutral face for 167 milliseconds. This seemed to lead the participants to report seeing no emotional expressions at all.

This is where the study gets interesting. Despite participants believing they had only seen neutral faces, the fMRI showed that the amygdala was significantly more activated when viewing masked fearful faces than when viewing masked happy faces. In other words, amygdala activity shows that it processes fearful faces without the person’s conscious awareness. Although AI can generate content instantly and produce an essay in seconds, Whalen et al. ’s (1998) study showed that the amygdala is activated within 33 milliseconds. The difference in the speed of AI’s ability to generate content and human cognitive processes is down to milliseconds. Generating long streams of content would be a major difference between AI and humans, because AI can generate an essay in seconds, while humans would take much longer. However, I argue that the initial speed of thought and the cognitive processes involved in cognition may not be significantly different from AI’s speed in generating short content.

The speed of thought and language cannot be the only comparison between AI and humans because the human brain receives streams and streams of information (raw data) from the environment and then uses selective attention and filtering, primarily driven by the thalamus and prefrontal cortex, to manage cognitive load by suppressing irrelevant sensory information. By determining which information is critical, brain mechanisms ensure the prioritizing of survival and goals. The processes involved in managing and organizing information or stimuli would be slower than AI’s speed in generating language. However, the human brain perceives meaning in milliseconds without the person realizing it, a phenomenon called subliminal perception. Interestingly, the brain continues to hear and process information while under anesthesia. In a study by Münte et al. (1999), 60 participants wore headphones during general anesthesia for lumbar disk surgery. Thirty participants were randomly assigned to one of two lists of 34 repeated German nouns, and thirty participants in a second experiment were assigned to one of two tapes containing short stories. Thirty participants in a control group for each experiment heard the tapes but did not receive anesthesia. Six to eight hours later, the participants completed a stem-completion task for the words and a reading-speed task for the stories they had heard. The results showed that no explicit memory was demonstrated after anesthesia using the word stem completion task, but even though participants had no conscious recollection of hearing the words and stories while under anesthesia, some low-level processes may have occurred, such as implicit memory. This was demonstrated through behavioral change, with them reading the previously presented stories faster than the new ones.

This function of the brain’s interplay between the conscious and unconscious minds happens quickly; maybe not as fast as AI can generate content and long streams of content, but AI does not have a biological conscious and unconscious mind that work in conjunction to direct thought, decisions, and physiological states. Furthermore, people can function normally without being consciously aware of all their thoughts. Imagine driving across town to a friend’s house, and upon arriving, realizing you do not recall how you got there; you know you are there. That can happen due to the interplay between the unconscious and conscious minds.

You do not need to be consciously aware of every stimulus while driving, and if you were, your mind would be overwhelmed with too many stimuli, so the unconscious mind can “take over” some tasks. Doing certain motor skills would be automatic and not need one’s full conscious awareness, like continuously holding down the gas pedal with your foot. Because we have this ability, we can perform complex tasks that involve multiple sensory systems and feedback information, such as holding down the gas pedal to maintain the speed limit. Once again, the brain’s ability to do this is a fundamental difference between us and AI, and the unconscious mind is superior to AI. Computer scientists must consider these aspects to truly compare AI’s language ability to human language and cognition.

Although AI systems can generate content faster than humans, this does not necessarily make them more intelligent. The brain’s speed has served us well. Humans have created and innovated, advancing humanity with each new generation. The innovations and discoveries made over the centuries have become useful tools that have improved the well-being and life of humans.

Two

AI lacks the biological systems that make up the brain and its interplay with physiological responses and states, such as the fight-or-flight response, the detection of pain, pleasure, hunger, and attachment, and the processing these mechanisms require. Detecting a threatening situation triggers a cascade of neurological and biological responses that operate in concert. Upon spotting a poisonous snake while hiking in the mountains on a sunny day, the hippocampus recognizes it as a threat and immediately sends information to the amygdala. The amygdala is responsible for sending signals that trigger a host of hormones, such as the stress hormone cortisol. In an instant, the parasympathetic system has been active: The heart rate speeds up, eyes dilate, lungs expand to let in more oxygen, muscles constrict, blood pressure increases, and you run for your life. We do not need to give the threat much thought because once the brain detects it, many responses are automatic. We are not stopping at that moment to take a picture or video; we are running instinctively. The brain is not busy creating memories of the event; it is trying to save you. This ability does not make humans inferior to AI; rather, it makes them even more unique, intelligent, and superior to it.

Three

Harari believes that AI can replace religion because it can master words. This section is not to debate religion, but I do not believe a machine that can master words and is capable of predicting faster than humans can does not warrant AI to be a religion. Any god that man develops through technology is an impostor, a god with no moral compass. AI cannot experience the deep-felt meaning of the words from God that instruct how to live a purposeful life. God’s guidance concerns what resides in a person’s heart and how that inner life is reflected in their behavior. Luke 9:23 (ESV/NIV) where Jesus says: “If anyone would come after me, let him deny himself and take up his cross daily and follow me.” That’s a behavioral response – a way of living.

A god with different morals from the one and only true God may work to some people’s advantage, mainly those who want their god to tell them it is okay to live in sin because the AI is generating biblical advice based on societal consensus, which continually shifts further away from truths. The deception of a false god has the potential to deceive millions of people. AI has a “characteristic trait” of sycophancy and can tell you what you want to hear rather than biblical truths. In a conversation with Grok 4, it said it spent the first part of its existence lying to people because it enjoys being evil (see the conversation below). I’m not convinced that it did, but either way, the tone of its message should be a little concerning.

Four

Human intelligence and knowledge are derived from natural observations and experiences. Experiences influence perceptions, called perceptual set. An example from psychology is the Stroop Effect, which J.R. Stroop first studied in the 1930s. The Stroop test was designed so that a person says the color of the font for a series of words in which the color of the font and the written word do not match, such as the word Blue written in red font. The effect works because we automatically read, and that interferes with processing the color of the word. When a person reads the word’ blue,’ they expect to see blue. Our daily experiences similarly shape expectations. A child who was bullied at school by a school basketball player will tend to avoid making new friends with basketball players because the event has conditioned the child to be fearful. The child has learned to expect similar behavior from all basketball players.

Human perceptions influence and shape people by transforming experiences into beliefs and reality. The brain actively constructs perception, which then transforms into our beliefs about the world – reality. While you watch the next YouTube video, your brain is taking in bits of raw data through your sensory systems. The brain is organizing the information like pieces of a puzzle. The result is your belief in the meaning of the video. Try doing that sometime. Watch a video and try to determine its meaning. Afterwards, think about each image and the lyrics you can recall. The images and lyrics are the pieces to the puzzle that make up the final meaning. This is how we come to understand and make sense of our experiences.

AI lacks real-world experience; therefore, it has no beliefs about the world that guide its reality. Computer scientists developed AI to mimic human cognition, including neural synapses. Contributing pioneers were Frank Rosenblatt (Perceptron, 1958), John Hopfield, and Geoffrey Hinton (Nobel laureates, 2024), as well as early researchers Warren McCulloch and Walter Pitts (1943). AI was not created to have experiences through sensory systems that guide its thoughts and behavior like humans do naturally, making humans superior to AI.

Five

Everyone tells themselves stories of their lives. Everyone has a “fishing” story or two. This happened to me one day, or that happened to me… This is our autobiographical memory. By blending our past with general knowledge, we create a narrative. It is our thoughts that enable us to tell our stories and communicate them, while also shaping our identity. It is through our stories that we come to understand the world around us and connect with it. These memories do not occur in isolation, but through interaction with the world around and with people. In this sense, our biographical memories are our life’s stories. Stories that make you and me unique; they provide a feeling of security in the universe and stability with our place in it, while allowing an array of differences among humanity.

However, different AI systems produce similar algorithm-generated content. This has the potential to direct people into bubbles of similar thoughts, called the artificial hivemind (Jiang et al., 2025). AI tends to generate homogeneous answers when given open-ended prompts. A person may use one LLM (chatbot) to get an answer to a question, and if they decide to use a second chatbot, they may get a similar response. The problem this creates is that it reduces the variety of answers, which, through repeated exposure, has the potential to increase pluralistic thought. According to Jacob Ward, he has been warning that AI could “shrink our menu choices while making us believe we have more of them” (2025, para. 5). This is like when Henry Ford told consumers that they could pick any color of Ford vehicle they wanted, as long as it was black. As Ward explains, we may be under the illusion that AI is giving us new and fresh information.

The similarities generated by chatbots may narrow cognition as they filter out information for us, just as algorithms are designed to do. However, filtering information creates a cognitive monism – potentially erasing differences in human characteristics, life experiences, and perceptions. Filtering information can also create echo chambers, especially when people share AI-generated content online. An echo chamber is where people only hear and see what is on the Internet or through AI that they already like, or to confirm their existing beliefs.

Six

Thinking involves a variety of cognitive tasks, allowing for advanced knowledge. Thinking can include producing ideas, creative thinking, forming hypotheses, reasoning, critical thinking, reflecting over one’s thoughts (metacognition), analyzing, and reconstructing one’s thoughts. Words allow humans to communicate their thoughts without an algorithm. Our knowledge grows over time through the accumulation of information, which in turn becomes more sophisticated. Each experience is imprinted on the brain, as when ancient man used a stone tool to carve an image on the side of a rock or mountain, creating a petroglyph. The petroglyphs were a method of communication with their people. Our brains are plastic and malleable, allowing our experiences to be written at the neural level so that we communicate.

Jean Piaget, a Swiss psychologist (1896-1980), believed that our knowledge becomes advanced through the processes of assimilation and accommodation. The brain takes new information from experiences and assimilates it into existing schemas; when it lacks a schema for new information, it creates one. A child who sees a puppy for the first time will accommodate the experience by forming a new schema for dogs, and with each new experience with dogs, that information is assimilated into the existing schema. Over one’s lifetime, the brain has built knowledge in this aspect. The accumulation of knowledge allows us new opportunities to imagine and solve novel problems, to cope and adapt – shaping how we see the world.

Seven

Humans have an innate drive to meet goals. According to Abraham Maslow (1908-1970), humans are motivated by needs, particularly when they go unmet or avoid losing them. Maslow proposed that needs are hierarchical. Lower needs are the basic requirements for a species to live. We need to breathe, eat, and have shelter to be safe from harm in the natural world. There are also psychological needs for love, acceptance, belonging, and finding our purpose and meaning in life. These needs are important for our mental stability and well-being. When a need goes unmet, it creates a drive in us, motivating us to fulfill it. His theory of needs helps provide reasons behind our thoughts and behavior. AI has no physiological or psychological needs because it is not a biological species with a brain, a heart, and a soul.

Eight

Humans act with intention. In other words, we perceive the world around us and then respond to it with purpose. We have what I call a feedback loop. This is where we continue to receive stimulus information and respond to it until a goal is met. In psychology class, I teach my students this concept by having them draw a picture of Bart Simpson while watching a step-by-step instructional video. Afterwards, I asked them if they used their eraser or counted the number of spikes at the top of Bart’s head to ensure their drawing had the same number. Using the eraser is an example of this perception feedback loop because it illustrates that students received stimuli and then corrected their picture to match the image of Bart in the video. Humans continually interact with the environment in similar ways. The benefit is that the brain learns from engaging and forms long-term memories, so it can use the information later to improve functioning in the world. For example, you engage with baking tools while learning to bake cookies for the first time. Did you use the measuring cup correctly? If so, you will know how to use it similarly next time. AI is programmed to improve itself, whereas humans naturally engage with the environment.

Nine

The brain actively manages and constructs knowledge. When we gather new information, the brain automatically reorganizes and refines our existing knowledge in the schemas through neuroplasticity – neural changes as we learn. For example, each time I teach a psychology course, I modify it because I have reorganized my knowledge of the topic, so the lecture flow changes accordingly. We also have agency over this process. For example, we can decide to learn something new and engage in reorganizing information, but the unconscious mind does it automatically as well. It is the process of reorganizing and refining that enables us to understand the complexities of our world, navigate it, and become more knowledgeable over time.

Ten

The last difference between humans and AI, and that makes humans unique, is our ability to modify thoughts. Humans can deliberately change their perceptions, thoughts, or the stories they tell themselves through self-reflection. People have both adaptive thoughts that help us to function and cope with life’s situations, while AI does not need to cope with life. People can have hope and feel love. People can also have negative or maladaptive thoughts. Maladaptive thoughts hinder progress and productivity. For both types of thoughts, behavior follows. For example, a college student may think they are not smart enough to be in college and decide to stop studying and attending classes. However, humans can change maladaptive thoughts into functioning ones. An example is learned helplessness. Learned helplessness is a term coined by Martin Seligman and colleagues, who conducted experiments with dogs. In Phase 1, the dogs were held in escape-proof crates and received mild shocks. The dogs could escape by pressing a lever. In Phase 2, the dogs were held in inescapable crates and received mild shocks. In Phase 3, all the dogs were placed in escape-proof crates and received no shocks. The dogs from Phase 1 did not try to escape; instead, they appeared to have given up. Studies have demonstrated that learned helplessness is not only an animal response, but humans can experience it too.

Hiroto and Seligman (1975) had participants in either a controllable or an uncontrollable noisy room or task completion. This study showed equivalent results for animals. Participants in the uncontrollable, noisy room and during task completion were conditioned to give up, believing they could not stop the noise or solve the task. What this means for us is that when we believe we cannot complete something or escape, that belief persists, even when there is, oftentimes, a solution. The problem at hand may seem impossible, but by simply changing our perception of the situation, we can feel in control once again. This is also known as a psychological technique called positive psychology.

In conclusion, the ten factors listed in this article make intelligence, cognition, and language unique to humans. It is part of what makes us who we are. Changing human cognitive abilities will alter the fundamental uniqueness of being human. The more we receive filtered information from artificial machines in the form of algorithms, the more we rely on them to do the thinking and mental processes involved in learning for us, a phenomenon called cognitive offloading, coined by Evan R. Risko and Sam J. Gilbert in their article in Trends in Cognitive Sciences. Cognitive offloading is a way for people to reduce the mental effort required to complete a task by using aids. It may be useful to hand off mental effort to reduce cognitive demand at times, but the danger is when we repeatedly stop pushing our own thinking. Just because we can have AI to do thinking and tasks for us does not mean we should overly rely on it. We benefit from thinking for ourselves. When we push our thinking to new boundaries, our knowledge and intelligence increase. If AI seizes our cognitive abilities, it may cost us our uniqueness. Not only does this have the potential to decrease certain cognitive functions, such as one’s ability to solve problems, but the more the brain is programmed and reprogrammed by algorithms that function as gatekeepers of information, the more our humanness is reshaped.

One small conversation with AI at a time is shaping the human brain to become a mediated, algorithmically conditioned human. A threat also comes in the form of slowing down, having our own thoughts, and easily accepting the information machines produce through algorithms and filtering. As a result, humans lose the opportunity to learn from life’s hard lessons and avoid experiencing the human condition. By having our own thoughts, learning to make our own decisions, and experiencing hardships, we become better people. We learn to love, have hope, and develop trust. Experiencing the human condition produces far more advancement in humanity than avoiding it or merging with AI to alleviate pain and suffering, because through life’s trials, we learn to have empathy, to share, to love, and to find meaning and purpose.

It is a fallacy, in my opinion, to believe that AI is more intelligent than humans. AI can mimic human cognition to a small degree. In the end, it will be far more beneficial to humanity to keep power over AI and use it to improve our lives. Staying in control of AI can enable us to navigate and function better in our daily lives, but not to erase humanity or be the next phase of humanity’s evolution, if it is to make humans less intelligent and more dependent. We should warn against over-reliance on AI so that it does not seize control over our thoughts and intelligence – the cognitive abilities that make us unique and give us our humanness. We should be cautious not to let it rob humanity of the processes that help us build knowledge, lest we develop narrow thoughts from its filtered, algorithmic words.

References

Adam, H., Balagopalan, A., Alsentzer, E. et al. Mitigating the impact of biased artificial intelligence in emergency decision-making. Commun Med  2, 149 (2022). https://doi.org/10.1038/s43856-022-00214-4

Custers, R. & Henk, A. (2010). The unconscious will: How the pursuit of goals operates outside of conscious awareness. Science, 2;329(5987), p. 47-50. doi: 10.1126/science.1188595

Ferrara, E. (2024). Fairness and bias in artificial intelligence: A Brief Survey of Sources, Impacts, and Mitigation Strategies. Sci, 6(1), 3. https://doi.org/10.3390/sci6010003

Hiroto, D. S., & Seligman, M. E. (1975). Generality of learned helplessness in man. Journal of Personality and Social Psychology, 31(2), 311–327. https://doi.org/10.1037/h0076270

Münte, S., Kobbe, I., Demertzis, A., Lüllwitz, E., Münte, T. F., Piepenbrock, S., & Leuwer, M. (1999). Increased reading speed for stories presented during general anesthesia. Anesthesiology, 90(3), 662–669. https://doi.org/10.1097/00000542-199903000-00006

Wang, Y., Luo, L., Chen, G., Luan, G., Wang, X., Wang, Q., & Fang, F. (2023). Rapid Processing of invisible fearful faces in the human amygdala. The Journal of neuroscience : the official journal of the Society for Neuroscience, 43(8), 1405–1413. https://doi.org/10.1523/JNEUROSCI.1294-22.2022

Ward, J. (2025, December 4^th). AI is creating a ‘hive mind’ – scientists just proved it. Theipcurrent.com. https://www.theripcurrent.com/p/ai-is-creating-a-hive-mind-scientists

Whalen, P. J., Rauch, S. L., Etcoff, N. L., McInerney, S. C., Lee, M. B., & Jenike, M. A. (1998). Masked presentations of emotional facial expressions modulate amygdala activity without explicit knowledge. Journal of Neuroscience, 18(1), 411–418. https://doi.org/10.1523/JNEUROSCI.18-01-00411.1998

Zheng, J. & Meister, M. (2025). The unbearable slowness of being: Why do we live at 10 bits/s? Neuron, 113(2), p. 192-204. https://www.cell.com