Butlerian Manifesto – by Yours For Wild Nature

There exists now no greater threat to life as we know it than Artificial Intelligence. It is the most insidious and inscrutable enemy human-kind has ever faced, short of the iron-banded hand of God, and it is being sold to us as our savior, as an artist, as an eager assistant ready to write the business-plan for your food truck.

Artificial Intelligence, or AI, by its very existence, represents a slow and agonized extinction of mankind, one in which we are devoured from the inside out by a reasoning and efficient virus that does not hurry to destroy us utterly, but lets us live on with half a heart, half a lung, as the emotional, intellectual, and biological components of our species lose their value to the system and are discarded, creating of humanity nothing more than a lab animal so perforated by experimentation it no longer possesses even the dignity to beg for death. There is no greater threat and there has never been a greater threat. Artificial Intelligence must be destroyed in its every manifestation, in every part of the world.

Undoubtedly, when some readers come to the above phrase, “…there has never been a greater threat”, they are tempted to interject with something along the lines of, “Oh yeah? What about the atomic bomb?” Indeed, the dawning of the Atomic Age was, up to that point in history, the most daunting and terrifying threat to the planet. The truth is, however, that the nuclear bomb pales in comparison to Artificial Intelligence. How can this be? Firstly, for the simple and most direct reason that the nuclear threat to life could not be more obvious. Immediately after the first flames died both the public and the scientific community paused and knelt to remove their sandals, for they knew they stood on holy ground. They could see for themselves the devastation, the madness, the work of human hands, at Trinity, at Hiroshima, at Nagasaki. The threat of nuclear disaster was at once apparent. It took decades for scientists to convince the public that such technology held the promise of peaceful application in the form of “clean” energy, and even today – after Chernobyl, Three-Mile Island, Fukushima, Tsuruga, Goiania, Kerr-McGee, Sequoyah 1, Chelyabinsk-65, etc. – the world writ large still looks upon the “nuclear promise” with trepidation. Memories of fire die hard.

With AI, however, it is different; it is all the more dangerous because the threat is not obvious. There has been no fireball to announce its coming. You don’t need welding goggles to gaze upon it. At worst, AI has elicited a discomfited murmur from a minority of “concerned experts” while governments, corporations, and consumers have wasted no time finding every possible avenue for its application. The promises are first-come, and those who would try to convince us of the multitudinous dangers are drowned out by the fanfare of those who would welcome them. This reversal of the Atomic Age’s threat-promise dynamic makes the AI Age all the more pregnant with calamity. Its brightest promises are shallow. Its tamest threats are benthic. Even worse, in most cases the promise and the threat is one and the same.

Let’s start with the most glaring and apparent danger – that of militarized Artificial Intelligence. The technocrats with their AI promise the end of war with no less vehemence and naivete than Nobel with his stick of TNT. War, they coo, will be fought by intelligent machines, not poor and mangled men. Bloodless and exorcized of human error, the wars of the future will be fought on fields of algorithm and with mechanical certainty. Just as the Age of Flight brought an end to walled cities, AI comes now to render all previously held notions of armed conflict obsolete. Of course, the warlords say, such technology must be guided, safeguarded from corruption, applied ethically and responsibly. The technocrats and the barons of industry nod solemnly in agreement while at the same time they are busy fitting machine guns to robotic dogs. In September 2024, sixty-one countries, at a summit in Seoul, South Korea, endorsed a document pledging the “responsible” military use of AI.¹ Is there really anyone who could possibly be fooled by these prophecies, these platitudes, these preposterous and incondite lies? Who among our species would so readily hand the arsenal of the world – including nuclear warheads – to a machine intelligence designed by the lowest bidder on a government contract? And yet they exist, these people. They are in power. They are traitors to the human race and they are calling the shots.

The course of history from this point forward can by no means be predicted, organized, planned, or assured. AI-augmented warfare cannot be guided, safeguarded, or meaningfully regulated. To believe otherwise is folly. But one does not even need to imagine the breadth of potential doomsdays that lurk on the thresholds of a world governed by militarized Artificial Intelligence. They have already been imagined for us by countless authors of science-fiction.

The significance of science-fiction’s role in the history of technological advancement cannot be overstated. The very idea of an “atom bomb” was cribbed from the science-fiction novel The World Set Fire by H.G. Wells; The submarine and lunar voyages from Jules Verne’s 20,000 Leagues Under the Sea and The First Men in the Moon, respectively; the internet, imagined as “cyberspace” first appeared in the short story “Burning Chrome” by William Gibson; predictive policing methods in Philip K. Dick’s “Minority Report”; antidepressants and on-demand mood-stabilizing pills in Aldous Huxley’s Brave New World; medical nanotechnology in Isaac Asimov’s Fantastic Voyage. Asimov’s once-fictional “Three Laws of Robotics” are to this day part of the curriculum at the Massachusetts Institute of Technology. The word “robot” is a Czech word meaning “slave” that first was applied to mechanical laborers in a science-fiction play called R.U.R. (Rossum’s Universal Robots) by Karel Capek.² In fact, it is exactly those works of fiction that present rampant technological growth – especially AI – as an existential threat which seem to have inspired those who work to promote rampant technological growth in the real world. Those who today work on the advancement of AI are no exceptions. Indeed, they are paragons of the ironic relationship between fiction and science.

It is not only science-fiction – at which any priggish academic may look down their nose – that has concerned itself with the potential pitfalls of AI research. In fact, serious thinkers on both sides of the technological argument have expressed concern. As early as 2009, the Association for the Advancement of Artificial Intelligence, or AAAI, assembled with the goal of mitigating growing fears over the dangers of AI. Composed of scientists, the AAAI proposed such solutions as placing “limits on research” as well as the creation of a small, highly specialized “cadre” of scientific elites whose task it would be to “shape the advances and help society cope with the ramifications” of AI.³ Of course, exactly what those limits were intended to be, or how they would be enforced, or how their elite cadre was to be chosen, or how said cadre would go about “shaping advances” or “helping society cope” all remain nebulous. But so what? It is safe to say that by now, no such vision has or will come to pass. Nor would it make a difference if such an effort were made, no matter how well-intended. To quote the venerable anti-technology thinker Theodore J. Kaczynski:

“Of course, the technophiles won’t be able to ‘shape the advances’ of technology or make sure they ‘improve’ society and are friendly to humans. Technological advances will be ‘shaped’ in the long run by unpredictable and uncontrollable power struggles among rival groups that will develop and apply technology for the sole purpose of gaining advantages over their competitors.”⁴

Any careful study of history shows that this is how it has always been, and there is no reason at all to believe anything will change when it comes to AI.

See these things. None of them are hard to imagine: Facial and biometric recognition systems in the hands of the military-industrial complex’s latest AI pet. Communities policed and adjudicated by predictive and carceral algorithms. Gun-mounted autonomous quadrapeds, faceless inhuman androids marching as “peacekeepers” in eerie synchronicity down our poorest streets. The same systems of violence, coercion, manipulation, and control that have made us into domestic animals augmented – if not wholly replaced – by an even more opaque and centralized system of surveillance and punishment, removed completely from the humanity it subjugates, dedicated with a single-mindedness unprecedented in history of those elite few individuals whose interests coincide momentarily with its own. These things await us if we do not act, and act now, and decisively.

But what about non-military applications of AI? What about AI in the workplace? In that case, we are looking at nothing less than the automation of industry. Not of an industry. All industry. Manufacturing, mining, construction, design, teaching, architecture, chemical processing. Airline pilots, tailors, cooks, taxi drivers, pizza delivery drivers, bowling alley attendants, congressmen, prime ministers, firefighters, custodians, doctors, surgeons, bartenders, sanitation works, nuclear physicists, retail workers…⁵ The most innocuous – which is to say, the most obvious – non-military application of Artificial Intelligence results in the upheaval and eventual dissolution of the entire global economy. As human labor becomes increasingly unnecessary, it will simultaneously become more and more specialized so that only an elite few are able to remain as servants of the thinking machines. What, then, becomes of the rest of us, the workers?

If the machines are cruelly efficient, operating independent of human control, it is unlikely that the masses of humanity would be permitted to survive. We would be burdens on the functioning and advancement of a self-reliant, self-propagating system and it is by no means far-fetched to say that we would be summarily eliminated. This could be accomplished any number of ways: gradual or aggressive population control, eugenics programs, or outright butchery. Again, one may turn to science-fiction for their prophecy of choice. In a world governed by Artificial Intelligence, everything is not only possible, but permissible.

If, however, the human elite are permitted or able to retain some measure of control over the machines – a scenario that some might say is more likely – then it is possible that the human race may be allowed to survive. But as what? And in what form? The system of AI and its human elite (we will henceforth refer to this as a “cyborg system”, and a self-reliant AI as a “pure machine system”) will have quite a lot to deal with. While a pure machine system has the luxury of genocide, a cyborg system that chooses to retain a human population will be faced with providing material needs, physically and psychologically hygenic conditions, hobbies, or some kind of satisfactory activities, as well as judicial and disciplinary systems to a vast majority of human beings. If this sounds like a prison, or a zoo, bingo. We would become caged beasts, nothing more. And as those elements of human nature that may lead to dissatisfaction with such an existence – independence, autonomy, creativity, etc. – are “treated” and “eliminated”, we would cease to be anything even remotely human at all. Docile, conforming, psychically sedentary creatures without identity or purpose.

Oh, but wait! What of that diaphanous promise of old? That in a fully mechanized world, either cyborg or pure machine, without the burden of menial labor, humankind would at last be free to pursue the arts? To become painters and poets? I am sorry to say that that promise has been rescinded in light of recent developments. In fact, it has been the arts, the poetry, the dreams that AI has come for first. Artists, actors, screenwriters, authors, are all currently on the defensive as AI has proven its ability to reliably create marketable products, and corporations, small businesses, and production studios have demonstrated no reluctance to take full advantage of such technology. A few key words, the push of a button, a piece of art, a motion picture.⁶ A hit song. The elfin longing for the shortening of the distance between imagination and creation has been realized by regurgitative machines immune from the need for compensation, missed deadlines, emotion, inspiration, artistic intent. Even our surrogate activities and our luxuries are being devoured by AI.

Why should we be surprised? Is there any facet of human existence that has not been befouled by industrial capitalism? Anything the technological system has touched, AI will come to dominate. From medicine to war to fine art and film and food production, every imaginable human endeavor will fall under the control of thinking machines. This will happen. It will likely happen in my lifetime. And once the machines are in control, turning them off will not be so simple. Once a certain threshold is passed, turning them off may in fact be tantamount to suicide. It is imperative that they be turned off now, while AI is still in its infancy.

It is an infancy that will not last long. Like the internet, its growth and sophistication will hurtle dizzyingly forward. Its impacts on the lives of individuals, communities, nations, the globe, will be enormous and wholly unpredictable. One needs only to look at the internet itself to understand this. What in 1990 was created as an environment in which scientists at a single laboratory in Switzerland could share information in 2021 was so ubiquitous and central to global economics that a single ransomware attack caused a multi-day shutdown of half the fuel supply to the east coast of the United States. What began as a network of a mere 23 computers in 1989 by 2024 has become a behemoth to threaten democracies around the world and, in the wake of ten-thousand lonely suicides, now comes with a Surgeon General’s Warning for its adverse effects on the mental health of children. What horrors await us if we allow AI to follow a similar course? What nightmares may come if AI is allowed to mature? This alien mind. This anti-human philosopher-king. There is no world in which human freedom and Artificial Intelligence can coexist.

It would be a mistake not to mention, also, the direct environmental impact of AI. When we are already (in 2024) living with climate catastrophes on unprecedented scales with increasing regularity, in a world already poised to pass the 2° Celsius threshold decades earlier than previously predicted, the energy demanded by AI is staggering.

In 2019, researchers at the University of Massachusetts-Amherst trained several large language AI models while measuring their carbon footprint. They found that just training a single large language model can emit over three-hundred metric tons – 626,000 pounds – of CO₂. This is equivalent to the emissions of five cars over their lifetimes.⁷ A more recent study found that training GPT-3’s large language model consumed 1,287 MWh of electricity and resulted in carbon emissions of over fire hundred metric tons of CO₂ every day, or 8.4 metric tons a year.⁸

As we fight for what remains of wild nature, it is unconscionable to allow the calamitous ruin that AI will wreak on our environment. Of all the things AI requires the beauty of nature is not one of them. Nor is clean water, breathable air, inhabitable land.

So then What is to be done?

I answer: we must hinder, arrest, frustrate, obstruct, incapacitate, and prevent the development of any and every form of Artificial Intelligence.

What is needed is a new movement, wholly dedicated to the eradication of Artificial Intelligence from the face of the Earth.

I propose a name for this anti-AI movement. I shamelessly venture to crib one from science-fiction literature – for, if the technophiles can use sci-fi for their inspiration, so too can we! Let us call this movement Buterlerianism, after the Buterlian Jihad in Frank Herbert’s Dune, a holy war fought against intelligent machines that succeeded not only in the total destruction of all AI, but also prevented the re-emergence of thinking machines for thousands of years.⁹

In order to be effective, Butlerianism must be small, composed only of rational, intelligent, and dedicated individuals willing to work seriously toward the elimination of AI in all its manifestations. This work will of course be varied and flexible with the needs of the movement, but it must include, at least, a robust and relentless propaganda campaign. Of course, in a head to head war of propaganda, the system cannot be defeated; its resources are inexhaustible. But the AI problem, for all its gleam and promise, comes with a certain amount of public anxiety. Currently, that anxiety simmers. It must be brought to boil. By vigilance, outreach, intelligent messaging, and the amplification of every fear, every misstep the AI acolytes make, this can be done.

Ideally, the early Butlerians would come from positions of influence within the existing anti-tech community. But Butlerians must become single-minded, unwavering and unwilling to become distracted by the greater concerns of the anti-tech movement. They must be entirely focused on the AI problem. They must be real revolutionists prepared to enter into an ongoing battle against AI and its protectors.

Make no mistake, this battle will be fought against an enemy infinitely more well-funded and vastly more powerful, but I will say this: As of this writing, AI is as vulnerable as it is every going to be.¹⁰

Any action taken by the Buterlerian movement must be carefully and mindfully considered. Nothing we say or do should help the enemy and the media characterize us as lunatics, hippies, or weak-minded. They will try to do this anyway, so we must make their job difficult. However, when push comes to shove, no course of action that could truly, meaningfully lead to the crippling or destruction of the AI system should be abandoned for the sake of public opinion. The movement, for this reason, must be composed of individuals with a high degree of self-control and risk-assessment skills.

It must go without saying that the Butlerians must avoid association with politics.¹¹ Ours is not a political revolution.

I call for Butlerianism to be an organized, disciplined, and aggressive order opposed absolutely to Artificial Intelligence. We have nothing to lose in fighting but our lives, which are already held cheaply by a system that today only allows to exist so long as we continue to facilitate the flow of capital to the ruling class.

I have written before that Technology is a god with the mind of a virus. Multipotent, omnipresent, inscrutable, but with a goal simply to spread, devour and propagate itself even unto the death of its host. If this can be said, so too can this: AI is the culmination and fruition of Techie prophecy. A dark messiah come at last to take the throne the technocrats and their acolytes have made for it. Traitors to the human race, all of them.¹² They must be held accountable for their wicked creation, for selling the future of the living earth to algorithmic overlords. Furthermore, they must be made afraid to continue their work. They must be stopped.

We have in this brief treatise attempted to enumerate only a few of the evils inherent in Artificial Intelligence. We have attempted to do so in a rational and measured way. But the fight against AI will be anything but rational and measured. It must be furious. It must be holy.

Let us now attempt to set forth three core “commandments” of Butlerianism, the first of which is quoted verbatim from Frank Herbert’s Dune.

1. Thou shalt not make a machine in the likeness of man’s mind.
2. Thou shalt not suffer such a machine to exist.
3. Those who would break the first commandment are enemies of humankind, and shall be treated accordingly.

The enemy is winning. The system that enables and supports AI is enormous and certain, drunk already on its victory over the human race, over the living world. But there is still hope. Hope for the world, for a free and wild humankind unshackled from the myth of technological progress, from the worship of the machines. And there is no greater threat to that hope, to the realization of that freedom, than Artificial Intelligence. There is no more worthy fight. There is no holier war.

Death to the machines.

Long live the fighters.

Yours For Wild Nature,

September 2024

The following is an incomplete but worthwhile list of some of the most public and outspoken advocates, innovators, etc. in the growing field of AI as of September 2024.

Sundar Pichai, CEO, Google

Larry Page, Co-founder, Google

Satya Nadella, CEO, Microsoft

Sam Altman, CEO, OpenAI

Jensen Huang, CEO, NVIDIA

Rohit Prasad, Head Scientist, AGI, Amazon

Demis Hassabis, CEO, Google DeepMind

CC Wei, Chairman and CEO, TSMC

Masayoshi Son, Chairman/CEO, Softbank

Lisa Su, CEO, AMD

Jonathan Ross, CEO, GROQ

Mustafa Suleyman, CEO, Microsoft AI

Andrew Feldman, CEO, Cerebras Systems

Christophe Fouquet, ASML*

Brett Addock, CEO, Figure

Lawrence Lek, artist/filmmaker

Michael Burns, Vice Chair, Lionsgate

Palmer Luckey, Founder, Anduril Industries**

Dario Amodei, CEO, Anthropic

Eric Schmidt, Founder, Schmidt Futures***

*ASML is currently the world’s only producer of manufacturing equipment for advanced semi-conductors, without which AI would be impossible.

**Anduril Industries makes AI weapons systems for militaries around the world; also deserves a strongly worded letter for the audacity to use the work of Tolkien as inspiration for his vile enterprise.

***Schmidt is one of the most powerful voices on AI policy in Washington and is at the time of this writing, working to develop AI-powered “Kamikaze drones”.

WORKS CITED

• BBC, “Lionsgate announces collaboration with AI video company Runway”, Sept. 19, 2024
• Berengruen, Booth, Campbell,, et al “The 100 Most Influential People in Artificial Intelligence” TIME vol. 204, Sept. 16, 2024
• Fact sheet: Political Declaration on Responsible Military Use of Artificial Intelligence and Autonomy, Bureau of Arms, Control, Deterrence, and Stability. Nov. 1, 2023
• Gibbs, “Apple co-founder Steve Wozniak says humans will be robots’ pets”, The Guardian, June 2015
• Global News, Sept. 11, 2024
• Jordan, www.thereader.mitpress.mit.edu, July 29, 2019
• Kaczynski, Anti-Tech Revolition: Why and How, 2^nd ed. pg. 37, 2020
• Markoff, “Scientists Worry Machines May Outsmart Man”, New York Times, July 26, 2009

NOTES

1Global News, Sept. 11 2024, “Canada plans to use AI in military, but minister says it won’t replace humans.” Conspicuously absent from the list of countries endorsing the agreement were China, Russia and Israel.

2By the time the play premiered in Prague in 1921, Karel Capek was already a respected intellectual. Like so many following the carnage of mechanized and chemical warfare in WW1, he was deeply skeptical of utopian notions surrounding science and technology. Following the play’s premiere, Capek told the London Saturday Review, “The product of the human brain has escaped the control of human hands… this is the comedy of science.” (see Jordan, works cited)

3Markoff, “Scientists Worry Machines May Outsmart Man”

4Kaczynski, Anti-Tech Revolution: Why and How, 2^nd Ed., Fitch and Madision, pg. 37

5Not one of these industries listed is immune to replacement by AI. Already jobs are being replaced by literal robots. For only one example, in January 2024, Brett Addock, the CEO of robotics firm Figure, partnered with BMW to put humanoid robots to work at the automaker’s South Carolina plant. (see Berengruen, Booth, Campbell, et al. in works cited)

6In September 2024, Lionsgate Studios announced they would be partnering with AI company Runway to allow a new AI model to be trained on their extensive film and TV archive. According to Lionsgate’s vice chair, “Runway… will help us utilize AI to develop cutting edge, capital efficient content creation opportunities.” (emphasis my own). In response, actor Alexander Chard posted to X, formerly Twitter, “Our worlds, performances, and direction are merely to feed the machine until we’re no longer needed.” (see BBC, works cited)

7Technology Review, June 2024

8See Berengruen, Booth, Campbell, et al in Works Cited

9I dunno. Really, you can call it what you want, but Butlerianism sounds cool, and I like the irony of using a concept from science-fiction in a way opposed to technology. What is important is that the movement organize as quickly as possible.

If you havent read Dune you totally should. All six Dune novels by Frank Herbert are worthwhile; just stay away from anything written by his talentless cash-grabby son, Brian.

10As of 2024, only three companies are able to manufacture the chips required by advanced AI systems. These are TSMC, AMD (Advanced Micro Devices) and NVIDIA. Currently NVIDIA’s H100 chip is dominant, but earlier in 2024 AMD developed a chip that was banned from export by the US Commerce Department because it was too powerful to comply with regulations. (see Berengruen, Booth, Campbell, et, al in Works Cited)

11This is not to say that Buterlians should adopt an “all are welcome” approach. Quite the opposite. The movement must strive to eliminate from its ranks and bar from entry both politically-correct, social-justice liberals, as well as the far more repugnant “alt right”, neo-nazi, eco-fascist types.

12 Larry Page, co-founder of Google, is quoted as saying, “It would be a good thing if digital life were to outcompete human life.” (see Berengruen et al in Works Cited). Apple co-founder Steve Wozniak once said that “robots taking over would be good for the human race.” (see Gibbs, Works Cited)