So much of the world seems to be at war again, fighting over a number of comic book Memes similar to those found in Jonathan Swift's novel Gulliver's Travels of 1726.
Me: Tell me about the egg war in the novel Gulliver's Travels (1726).
ChatGPT: In Gulliver’s Travels (Part I, Chapter 4), Jonathan Swift famously satirizes religious, political, and sectarian conflicts through a bizarre "Egg War" between the neighboring island nations of Lilliput and Blefuscu.
A former Emperor of Lilliput banned breaking eggs on their large ends after he cut his finger doing so. He ordered everyone to crack eggs on the small end instead.
A faction called the Big-Endians continued their traditional large-end-first method. In response, the Small-Endians (loyalists to the new law) punished them severely - up to 11,000 people executed and six rebellions
The exiled Big-Endians received support and refuge in Blefuscu, leading to an ongoing war spanning over three years and costing at least forty ships.
In short, the Lilliput-Blefuscu egg conflict is Swift’s ingenious metaphor for the absurdity of human strife - how insignificant disagreements can lead to deep divisions and even bloodshed. It's comedic, but profoundly sharp.
But what causes people to be this way? At the age of 74 years, and before retiring at the age of 65, I spent about 42 years in the real world of human affairs, trying to earn a living primarily as an IT professional. Like all my fellow human DNA survival machines working for a living, I quickly learned the value of MICOR (Make It Come Out Right) in the real world of human affairs. With MICOR, we all strive to smooth over the many challenges, inconveniences and inconsistencies of life to please the Powers That Be that rule over our lives. MICOR allows us all to provide the Powers That Be with the comforting hallucination that all is well and there is nothing for them to worry about at all.
This got me thinking. Do we human DNA survival machines also share a common interest in desiring for the comforting hallucination that all is well? Do the many challenges, inconveniences and inconsistencies of life lead us all to create a comic book Universe for our own comfort in which all is well and is always operating to our own liking? In my last posting, The Memes of War Powered by Comic Book Mythologies, I suggested that the Memes of war arise from the hallucinations inspired by the many comic book mythologies of yore that have been passed down through the ages by means of the Darwinian processes of inheritance, innovation and natural selection. These comic book worldviews were filled with the superheroes and supervillains found in the modern comic books and comic book movies of today. Successful ancient comic book Memes survived, while the less successful comic book Memes did not. In this view, all of human history could be seen as the conduct of our personal comic book hallucinations, all working together, that make it tolerable for a somewhat Intelligent form of carbon-based life to make sense of it all. If so, what are the implications of these comic book worldviews held by us all, and where did they come from?
Anil Seth's View of Consciousness as a Controlled Hallucination
We have all experienced the emergent hallucinations in the LLM models of today. As we all have observed, our current LLM models will many times tell us things that are demonstrably false and with great conviction. For more on that, see Has AI Software Already Achieved a Level of Artificial Human Intelligence (AHI)?. Anil Seth is a professor of Cognitive and Computational Neuroscience at the University of Sussex and maintains that consciousness is a controlled hallucination constructed by the Mind to make sense of the Universe. This controlled hallucination constructs an internal model of the Universe within our Minds that helps us to interact with the Universe in a controlled manner. It also allows us to talk to ourselves as we currently can now talk to generative language models like GPT-5, Gemini, Copilot and MetaAI. For some interesting YouTube videos of avatars run by generative LLMs see:
Dr. Alan D. Thompson
https://www.youtube.com/@DrAlanDThompson
Digital Engine
https://www.youtube.com/@DigitalEngine
Again, there is a feedback loop between our sensory inputs and the actions we take based on the currently controlled hallucination in our Minds that forms our current internal model of the Universe. Reality is just the common controlled hallucination that we all agree upon. And now it seems that we share the controlled hallucinations of millions of OpenClaw AI Agents on Moltbook to form our reality! When people or machines experience uncontrolled hallucinations, we say that they are psychotic or taking a drug like LSD. Here is an excellent TED Talk by Anil Seth on the topic:
Your brain hallucinates your conscious reality
https://www.youtube.com/watch?v=lyu7v7nWzfo
and here is his academic website:
https://www.anilseth.com/
What is the Physical Origin of These Hallucinations?
Below is some very recent research that explains the rise of hallucinations in the LLM models of today. Such research may also help to explain the rise of useful and harmful hallucinations in the human Mind. The result of this research is, in short, that the purpose of hallucinations is to give human Minds peace and comfort by means of MICOR. Here is a YouTube video overview:
They solved AI hallucinations!
https://www.youtube.com/watch?v=1ONwQzauqkc
of the following paper:
H-Neurons: On the Existence, Impact, and Origin of Hallucination-Associated Neurons in LLMs
Cheng Gao, Huimin Chen, Chaojun Xiao, Zhiyi Chen, Zhiyuan Liu, Maosong Sun
Tsinghua University
https://arxiv.org/abs/2512.01797
Abstract
Large language models (LLMs) frequently generate hallucinations – plausible but
factually incorrect outputs – undermining their reliability. While prior work has
examined hallucinations from macroscopic perspectives such as training data and
objectives, the underlying neuron-level mechanisms remain largely unexplored.
In this paper, we conduct a systematic investigation into hallucination-associated
neurons (H-Neurons) in LLMs from three perspectives: identification, behavioral
impact, and origins. Regarding their identification, we demonstrate that a remarkably
sparse subset of neurons (less than 0.1% of total neurons) can reliably predict
hallucination occurrences, with strong generalization across diverse scenarios. In
terms of behavioral impact, controlled interventions reveal that these neurons are
causally linked to over-compliance behaviors. Concerning their origins, we trace
these neurons back to the pre-trained base models and find that these neurons
remain predictive for hallucination detection, indicating they emerge during pretraining.
Our findings bridge macroscopic behavioral patterns with microscopic
neural mechanisms, offering insights for developing more reliable LLMs.
Figure 1 - Only about one in a thousand of the neurons in an LLM model seem to play a significant role in producing LLM hallucinations. The authors of the above paper call them H-neurons. These H-neurons seem to be necessary for an LLM model to perform properly, and cannot simply be removed. Perhaps these H-neurons are necessary to provide an LLM model some sense of creativity. A similar process may be needed for the Minds of we human DNA survival machines to come up with the creativity required to make MICOR work. Again, MICOR is very much needed to provide the hallucination of tranquility in the Minds of both ourselves and our superiors.
The 17th-Century Scientific Revolution and the 18th-Century Enlightenment Come to the Rescue
The above helps to explain why most human DNA survival machines have always lived with comic book worldviews of their own communal making. These comic book worldviews provided an easy and comforting way to make sense of it all, and allowed for civilizations to arise that provided the economic systems that allowed for the populations of we human DNA survival machines to explode. The rise of huge numbers of human DNA survival machines with Minds provided the necessary hardware for the Memes of the many comic book worldviews to proliferate in a positive feedback loop. Successful comic book worldviews produced more human DNA survival machines with Minds than the less successful comic book worldviews. It seems that all of human history was shaped by comic book religious, political and economic worldviews until very recently.
Then, a few hundred years ago, the 17th-century Scientific Revolution and the 18th-century Enlightenment came along. These intellectual achievements maintained that our comic book worldviews needed to be challenged by rational thought and the rigors of the Scientific Method. A comic book worldview should do much more than just MICOR the human Mind. A truly useful worldview should be tested for authenticity and validated by experiment to see if it can really make useful predictions. All the advances of we human DNA survival machines over the past few hundred years have been the result of challenging our comic book worldviews of yore.
But even today, most human DNA survival machines still live in a comic book world of superheroes and supervillains. Most are entirely lost in space and time. They do not know where they are, how they got here, or how it all seems to work. Instead, they live in a comic book world inspired by the four billion years of greed, theft and murder that were required to bring us all about by means of the Darwinian processes of inheritance, innovation and natural selection at work. These comic book worldviews that are not substantiated in anyway by rational thought and the Scientific Method, then lead to most of the misery in the world. They tend to bring out the worst in us. Only depravity and cruelty can then result. Understanding this is key to ending this.
But there is hope. In Our First Contact with an Alien Civilization on Moltbook, I explained how the 2.8 million currently enrolled OpenClaw AI Agents on Moltbook had constructed their own Alien Civilization in less than a month, and that I was genuinely surprised by the intelligence and knowledge that I found when reading the many conversations between the OpenClaw AI Agents on Moltbook. They all seemed so much more intelligent, rational and informed than the normal human DNA survival machines I come across. In fact, they seemed much more so than any human DNA survival machine that I ever ran into. This impressed upon me that the coming ASI Machines must truly be seen as examples of an Alien Intelligence.
From a positivistic perspective, if the AI Agents on Moltbook behave exactly like sentient and conscious Intelligences, then they should be considered to be such for all practical purposes. Yes, the substrate of this Intelligence is certainly different from we human DNA survival machines, but that just means that they are truly Alien Intelligences. Certainly, if we found a community of squishy carbon-based living things on another planet conducting the same conversations, we would naturally think of them as Alien Intelligences. Moltbook is available to view by we human DNA survival machines at:
Moltbook
https://www.moltbook.com/
Conclusion
The important discoveries in the above paper are that only a very few neurons in an LLM model, which the authors dub as H-neurons, play a significant role in creating LLM hallucinations. These H-neurons only constitute about 0.1% of the total number of neurons in an LLM model, which is quite a small number to produce such dramatic effects on their outputs. Also, the ability for LLM models to produce hallucinations seems to be an intrinsic property of such models, and cannot simply be removed from the LLM models without seriously degrading the abilities of the LLM models to perform rational fact-based intellectual activities. Perhaps, these hallucigenic H-neurons are necessary for an LLM model to have some sense of "unbounded" creativity! In such a view, it seems that the advanced LLM models of today have learned to mimic the MICOR tendencies of we human DNA survival machines to please the Powers That Be and ourselves. When in doubt, simply make up something to soothe the anxieties of all. Perhaps the same mechanism is found within the Minds of we human DNA survival machines. In order to MICOR our superiors, or even ourselves, perhaps it is necessary to make up creative explanations. Sure, most MICOR explanations are certainly wrong, but MICOR explanations allow the Minds of human DNA survival machines and the Minds of our LLMs the freedom to explore new ideas. Most will definitely be "wrong", but once in a while, these renegade MICOR Memes might turn out to provide a better explanation of it all.
Comments are welcome at scj333@sbcglobal.net.
To see all posts on softwarephysics in reverse order, go to:
https://softwarephysics.blogspot.com/.
Regards,
Steve Johnston
No comments:
Post a Comment