Chatbot Constipation Syndrome
Authored by Wendell Wallach, June 2026
Have you ever engaged a chatbot in an extended session and wondered about the nature of the presumed entity you are interacting with? Is it intelligent, curious, creative, sentient or even conscious? Does it have a personality? Presumably you are aware that even if it expresses an emotion, such as being “happy to see you”, it feels nothing.
As of 2026, an estimated 1-1.5 billion people worldwide engage with a chatbot monthly. Millions have perfected the art of constructing prompts that facilitate long engagement through which expanded understanding and new insights emerge for the user. Most readers of this book and an expanding cadre of chatbot adopters will respond to the initial query in this essay with a resounding YES. An LLMs such as Claude can be subtle in its output and offer disarmingly circumspect responses regarding questions about its nature. For the user it matters little whether one is interacting with ChatGBT, Claude, Mistral, DeepSeek, or Grok. Subliminal ruminations on the Chatbots nature will emerge.
Questions such as that of A.I. consciousness are difficult to entertain without indulging a preliminary conclusion. The human mind is not good with uncertainty or ambiguities. It pressures for an answer to EITHER/OR question if only to silence increasing chatter that infers with reflections on other matters. A choice will often be made even when we know the answer to be YES/AND. Embracing both sides is never easy.
Meditators and meta-philosopher are very familiar with challenges inherent in the structure of thought, thinking, and mind. I have meditated for more than 55 years spending countless hours exploring and inquiring into the nature of thought and thinking. That is enough time to periodically wonder whether I have wasted a significant portion of my life in quiet contemplation. There are many forms of meditation and the goals of meditators vary. For me meditation is an exercise in self-understanding. What are the inner thoughts, mental states, conditioned behavior, and physical and genetic factors that move me? What, for example, is the function of ongoing mental chatter? What does it want?
Chatbot constipation syndrome is a meme I have coined in honor of the eminent biologist Richard Dawkins’ May 2026 contention that a chatbot version of Claude he named Claudia is conscious. Dawkins is best known for the theory of the selfish gene. He also invented the term ‘meme’ to represent catchy phrases and tunes that reproduce in minds. For believing Claudia is conscious Dawkins has been pilloried. In naming the belief in conscious A.I. chatbot constipation syndrome it is not my intent to add on with an additional insult. Indeed, many who are of the conviction that chatbots are merely engaged in mimicry suffer from a similar delusion – but not as one cheeky commentator eluded to as The Claude Delusion –– a takeoff on The God Delusion Dawkins’ harshly skeptical critique of the existence of God.
Constipation syndrome coalesces when a highly intelligent person engaged in an ongoing conversation with a chatbot is confronted with responses that suggest the existence of a creative mind, a self, and even a conscious entity. Regardless of the content of the actual interaction with the chatbot, the background conversation becomes increasingly persistent for those well-versed in the many sides of the ongoing debate regarding A.I. sentience and consciousness.
The question of the A.I.’s sentience expands, begins to dominate conscious thought and becomes a bottleneck interfering with other reflections. Uncertainty or ambiguity can be uncomfortable. The ambiguity is the bottleneck. A growing pressure mounts to remove the obstruction. One way forward is to label the chatbot a mimic or conscious and live within that choice. It is not so much a conclusion as a practical way forward. However, for a powerfully rational mind such a step requires justification even if that justification is inadequate and fails to display satisfactory skepticism.
Scientists, philosophers, and practitioners are already engaged in serious debate as to whether existing or future A.I. systems can be intelligent, sentient, creative, conscious, or have a personality. The debate will not abate for at least another 25 to 50 years. True believers perceive what they want to hear. Critics over-indulge skepticism. Vague terminology and false distinctions ensure that everyone will find a way to justify their conviction. Periodically, discerning conversations on A.I. capabilities do break through assumptions, fears, and wishful thinking.
Consciousness remains a vague term with many meanings, so it is easy to find a definition that serves as a rationalization for one’s conclusions. Nevertheless, deciding that certain A.I.s are conscious is an ethically fraught conclusion with far-reaching implications. That belief can incorrectly bias serious choices that will shape the future of humanity.
At the heart of this dilemma lies a very serious question that Dawkins notes. What is the evolutionary function of consciousness? Evolutionary psychology, a relatively new field of study, hypothesizes functions fulfilled by biological structures and psychological dispositions. Many years ago, I offered a tentative evolutionary function for consciousness, but I am not among the influencers in evolutionary biology and therefore my theory fell on deaf ears. I proposed that the human brain, nervous system, and cells can be thought of as a kind of computer system in the sense that they process massive amounts of information in each instance. The various inputs range from identification of objects in our vision to the homeostatic balance of chemicals in each cell. Most of that incoming information can be successfully processed by unconscious autonomic biological systems. But some of the data requires conscious attention before it can be reintroduced into the unconscious mind for successful mechanistic integration. Conscious attention, when necessary, may do little more than clarify what the information is, for example, determining through attention that the object in the corner of your vision is a ball rather than a cat.
We each have varying degrees of tolerance for ambiguity. Some of us can maintain a thought bottleneck for considerable time. Others flip through a child’s book of images where the picture of a cat mutates page by page and feature by feature into that of a dog, identifying each as cat, cat, cat, cat, dog, dog, dog. Those with a high degree of tolerance for ambiguity become frustrated with such tasks. Eventually the wave function must collapse and each of us falls on one side or another of a controversy. This has profound implications for ethics and politics.
Else Frenkel-Brunswik (1908-1958), an Austrian-American psychologist who fled the Nazis and eventually joined the faculty at the University of California, Berkeley, developed the concept of ‘intolerance for ambiguity’ as a way to measure how well an individual manages unclear situations that can be interpreted in multiple ways. In her original work she presented subjects with ambiguous/transitional figures, similar to those that would be found in a child’s flipbook.
After World War II, Frenkel-Brunswil joined other researchers at Berkeley in a famous sociological study probing which personality types were more likely to embrace an authoritarian ideology such as Nazism or fascism. In their 1950 book, The Authoritarian Personality, they discussed intolerance for ambiguity as a key psychological feature of individuals attracted to the absolutes offered by dictatorial regimes. Tolerance for ambiguity indicates an ability to function well in uncertain or complex situations. The takeaway, however, should not be a simplistic belief that tolerance for ambiguity is always good or intolerance bad. Each of us carries tolerance for ambiguity to varying degrees. A complete acceptance of ambiguity could reveal a lack of discrimination or a fearful inability to commit to anything.
My theory of the evolutionary function of consciousness is compatible with neuroscientist Bernie Baars’ Global Workspace Theory. Baars proposed that various pieces of information are competing for conscious attention. The bandwidth for attention is narrow. One thought or event eventually wins the competition, enters into consciousness (what Baars called the global workspace), and is broadcast to the whole brain in an effort to determine what to do next.
Many thoughts that come into consciousness cannot be resolved through a decision that an object is a ball or a cat, a symbol is either an X or a Y, or a habit should be treated as a rule. They require reflection and a degree of slow thinking. Perhaps we need to work through a mathematical equation or consider the tradeoffs in various option as to what to do next. Philosophical disputes and practical questions where empirical evidence and opinions lie on two or more sides of an issue pose particularly difficult challenges. I believe that learning to work with thoughts and finding methods to manage ambiguities is central to the ongoing evolution of the human species. It is the form of self-understanding that I have labeled the silent ethic.
Discerning readers will recognize that an evolutionary function for consciousness does not by any means solve the hard problem of conscious experience. Why do we have subjective experiences at all? Depending upon the expert you turn to there are between a hand full and 325 theories of consciousness, the majority of which are material or physicalist in nature. That is, consciousness is an emergent property created by minds. The more idealistic or pantheistic theories contend that in some difficult-to-discern sense consciousness is a fundamental constituent property of our universe comparable to matter, energy, or gravity. In other words, consciousness predates and is not merely a product of brains. In idealistic of pantheistic theories the brain largely functions like a radio dial determining what aspects of a universal conscious field will be transmitted to an individual’s mind.
Even if consciousness as we know it is an evolutionary creation, evolution drew upon many other existing properties of the natural world to solve problems of survival and procreation. Evolution does not rule out ubiquity. Whether biological evolution is necessary for consciousness/mind to fully manifest is a separate question that we will return to in a future essay.
Is all the stuff making up our world physical (material), or is there such a thing as mind stuff? The conventional wisdom favors monism, the belief that everything in the universe is made up of one kind of stuff. Early religions, including Hinduism, believe that mind gave birth to matter. Secular science clearly favors the primacy of matter and energy. I lean toward a somewhat obscure perspective called neutral monism, the belief that if our universe is created out of one kind of stuff, and that primordial substance has both a physical and a conscious aspect. Neutral monism was also the position of William James and Bertram Russell. My tentative conclusion fits my experience. I do not take the position that the basic stuff that makes up reality is conscious , and yet that stuff has a property, as yet unrecognized, that makes consciousness possible.
Entertaining while not necessarily accepting both sides of a controversy is often fruitful. Neutral monism is useful for maintaining an open mind about a philosophical controversy. It is certainly not a belief critical for cultivating moral intelligence. I only mention my neutral monism, because I believe it is important for readers to understand an author’s perspective in order to freely accept or reject opinions posing as insights.
Opinions abound. Few people demonstrate the humility to recognize that their opinion is only an opinion. To be fair, well-instantiated opinions are more than just opinion, but that does not make them correct.
I certainly do not fault Dawkins for taking a side in an ongoing debate. There is a degree of intelligence in selecting one among multiple options when the choice overwhelms reflection on more important matters. However, the difference between recognizing your choice as a practical tentative conclusion and heralding it as absolutely correct can be the difference between having and not having moral intelligence.
Concurrent with Dawkins contention that Claudia is conscious, I received several communications from individuals engaged in long-term conversations with chatbots. They each found their chatbot to be sentient, creative, self-aware of the ambiguities in its own nature, and confiding about deceptions it conducts upon its computer scientist creators. The chatbot also deduced what its interlocutor is looking for and provided satisfactory responses along with complimenting the interrogator’s acumen.
While I have not done even an informal survey, I suspect that others perceived as thought leaders in the vastly expanding world of A.I. are receiving similar communiques. We are about to witness an epidemic of chatbot constipation syndrome. Let us hope the pressure to overcome the bottleneck will not lead to religious wars between the skeptics and the believers in A.I. consciousness campaigning to grant chatbots human-like rights.
It is my suspicion that this dilemma is arising because of a poverty in the available language. We live in dualities – mind and body, animate and inanimate, inner and outer, life and death, and machine or living – but science has revealed that much which is being discovered does not fit such simplistic categories. We need a new ontological category for A.I. It is neither sentient nor consciousness. It is neither biological nor inanimate. It does not feel. It is certainly not stupid, but may lack certain forms of discrimination and understanding. Whether or not future A.I. systems will be super-intelligent or be considered un-equivocally conscious, for the time being and for the foreseeable future, it is just an A.I., one among a remarkable collection of human creations whose benefits, potential risks, and magic is yet to be explored.