The Way How Curiosity May Guide Science (And Us) To Avoid Nuclear Armageddon And Other Existential Threats

“Those who fight for freedom with the technologies of death end by living in fear of their own technology.” — Freeman Dyson [1]

Humanity more and more resembles to me a cat stuck in Edward Thorndike’s puzzle box. There’s no obvious path to escape, but, unlike cats in Thorndike’s experiments, we keep trying obvious — well trodden — paths. We abandoned “the method of trial and error, with accidental success” because we take our own wisdom too seriously. We believe that we already know all possible paths. Yet only the natural method of adaptation — epistemic foraging — can help us now in the same way as it helped Thorndike’s cats to escape from puzzle boxes more than a century ago.

Karl Friston proposed the emergence of curiosity as an existential imperative under the free energy principle [2,3]. Indeed, active inference [4] always begins with curious behavior; from simply moving our eyes to performing scientific (and social) experiments. From every perspective curiosity is key [5–9]. We are quintessentially curious creatures. Yet in my search for the path to escape the nuclear destruction I concluded that it isn’t enough to wonder; daring is at least equally important. We have to dare to act without knowing the outcome. Indeed, what is curiosity other than the imperative to discover “what would happen if we did that?”

Atomic Dreams

This story begins with a poem. Once upon a time in early March of 2022 a Ukrainian poet Polina Taranenko sent me a string of text messages that sounded like a poem. She spelled out her nightmare about her dying in a blast of the atomic bomb.

“I dreamed of such a sunny afternoon, a lot of people.. everything blooms.. and suddenly.. a wave of fire.. comes from the horizon

and there are many people in front of me

and they begin to turn around and burn in rows, like this … turned around — burned out, then the next row

and the wave reached me.. and it was strange to feel the heat in a dream, usually a dream is just a picture

and then I burned out and I felt it

and I woke up from the fact that I was gone

it wasn’t even that scary

fear arose later

and then I thought that you and I joke all the time about Sarah Connor

and I remembered that Terminator 2 begins with her dream..”

Polina was then in the Ukrainian town of Odessa. And I was very worried about her. I had a nightmare that was very similar to hers. I never told her, but she guessed. She became increasingly upset and afraid that her nightmare may soon come true. I tried to calm her by telling her that her dream was not a prediction, but a warning sent to us so that we could prevent the use of nuclear weapons in Ukraine or anywhere else.

Sometimes — in emotional excitement — we make promises which we have no idea how to fulfill. Then, we have to begin by working out the path from the end to the beginning — to move back to now, step by step from the goal that we seek to achieve, considering all obstacles and constraints. My friends in machine learning tell me this is called backwards induction, which is a mainstay of optimal planning and dynamic programming.

My past experience — in participating in international disarmament negotiations — told me that in an evolving situation, any political or civic efforts to eradicate nukes would be futile.

What if a team of scientists passionate about the goal of developing a new generation of technology — technologies of life — smart and humane, could find a solution to eradicate all nukes on Earth remotely and without the consent of such arms’ holders? Such a crazy idea came to my mind. It turned out that I wasn’t the only one.

In a Facebook conversation Luca Visinelli, an Associate Professor in physics at Shanghai Jiao Tong University, suggested “We should remove all nuclear weapons from the whole world. That’s the best solution.”

“This is certainly true. But how do you imagine the process of removal right now? Every dictator in the world is inspired now by Putin’s example. Even a group of terrorists can acquire a dirty bomb, since it is not expensive at all. So, what is your method for humankind to remove nuclear weapons? Through a magic wand?” Mikhail Shifman, a Professor of Theoretical Physics at University of Minnesota, responded to Luca.

For the last two years I have been working on a book about universal intelligence as an interplay of primary (active inference) and secondary (machine learning) methods of learning. Through this work I learned about outstanding scientists who conducted theoretical and experimental research in a wide array of fields: from theoretical physics and cosmology to neuroscience and experimental biology, but focusing on the same subject — universal intelligence. Unlike the scientific mainstream obsessed with the secondary learning method as the most modern and efficient, all were also researching the primary method of learning and applying it [to] themselves.

From the history of science, I knew that the primary method of learning was discovered by one of the founding fathers of modern psychology Edward Thorndike in 1898. Thorndike named it the “animal-like learning method” or “the method of trial and error, with accidental success.” He found this method of learning of relations between different stimuli “useless and antiquated” compared to stimulus-response learning [10] — the secondary method of learning that became the mainstream of psychology, neuroscience and artificial intelligence. However, another founding father of modern psychology and neuroscience Ivan Pavlov in 1933 concluded differently, “In Thorndike’s experiments, the animal becomes familiar with the relations between external things among themselves, with their connections. Therefore, it is the knowledge of the world. This is the embryo, the germ of science.” [11]

‘Miracles’ created by the primary method of learning in modern science turned my perception of what science can do upside down. We can find the right way by crossing out the wrong ways we tried. Existing knowledge can’t take us to the destination, but it can help us to winnow down the direction of search after each erroneous trial. This just is the scientist as a curious creature. And indeed, an apt description of the scientific process itself in the post-Popperian era.

Therefore, I wrote an email to Mikhail the next day, “Extraordinary problems require extraordinary solutions. If we need a magic wand to eradicate nukes, then we have to make the magic wand, first, and use it to eradicate nukes, second.”

By the time of sending that email I already received generally positive feedback to my initiative from some scientists profiled in my book and I began to believe that my crazy idea could get traction.

“I have thought a lot about this issue, and no magic wand is visible, even in principle. It is not clear where to start, from which side to approach it,” the response from Mikhail was both encouraging and disappointing.

On the one hand, it was encouraging because Mikhail showed real passion about the issue. On the other hand, it disappointed me because I thought that we already had an idea from which side to approach the problem.

The Magic Wand

“I think it is a good idea. I also think that to get traction you will have to be specific about the procedure for inventing solutions. And — to engage people — give them some indication of what these solutions look like.

The challenge that has to be faced is the following. Theoretical formulations of the kind that Vitaly [Vanchurin] has brought to the table, much like the theory of natural selection and the free energy principle, are great descriptions of the way things work. However, you cannot use the theory of natural selection to predict or identify the next adaptation; i.e., “solution”. One could identify key mechanisms that enable the “next adaptation”; for example, meta selection or selection for selectability. The corresponding mechanism for the free energy principle would be the emergence of curiosity (i.e., actively resolving uncertainty).

Perhaps this is what you had in mind when talking about a reality check — with nuclear disarmament as a use case?”

It was exactly what I had in mind but would never be able to frame as clearly as Karl Friston did. He — the most cited living neuroscientist and the creator of free energy principle [2,3] and active inference theory [4] — took the idea close to his heart from the very beginning. “I can remember nightmares about nuclear war in my childhood,” he wrote in response to my first email.

Following Friston’s suggestion I wrote an explanatory note that, according to Friston, did put “flesh on the idea”. It was accepted as a starting point also by Vitaly Vanchurin, a theoretical physicist and cosmologist, who created the universal theory of learning that claimed that the Universe itself is a neural network [12].

The core idea of ‘the magic wand’ was to leverage Friston’s and Vanchurin’s theories for launching a network of scientists that will leverage primary learning to discover nonexistent solutions to problems unsolvable from the point of view of secondary learning. The eradication of nukes will be the first real life use case for it.

I even made a fantasy example of how such a ‘magic’ solution might look like.

“A primitive electromagnetic lifeform much simpler even than a virus, might be created to resolve this problem. It will have spores: microscopic networks of phase states in phase space. Spores will spread throughout the Earth’s electromagnetic field and other fields and penetrate everywhere. They will look for the presence of nuclear graphite against the background of radiation from weapons-grade uranium and plutonium. When spores find nuclear graphite irradiated by weapon-grade isotopes, they will come to life, start feeding on pure graphite, multiplying and clogging the graphite with the products of their homeostasis. These products will absorb neutrons rather than repel them like graphite. So, when the graphite becomes sufficiently clogged with them, the chain reaction in the bomb becomes impossible. And our little antibodies will again turn into spores and will stay on guard, if suddenly someone else will try again to create the a-bomb.”

A bit later Michael Levin, an experimental biologist, also considered the ‘magic wand’ idea worthwhile pursuing. Levin is known for his mind blowing experiments with electro-morphogenesis: xenobots [13,14], two headed planaria [15], regeneration of frog’s limb [16]. In his work Levin adds an experimentalist’s layer to Friston’s theoretical foundation. His ‘minds everywhere’ experimentally inspired framework [17] could be also used for morphogenesis of human networks. Humans too may evolve into their roles spontaneously like stem cells differentiate — just because Levin’s framework is universal.

Bombs & Poetry

After a reasonably fruitful first month or so I began to receive only silence, skepticism or despair in response to my initiative.

Theoretical physicist, author and political activist Carlo Rovelli wrote to me: “I proposed the idea of considering global negotiations for reducing weapons, last fall; I got 60 Nobel prizes supporting the idea and signing the appeal, plus the Dalai Lama plus the Pope, and not a single media in the US covered the news. With great pain, I am realizing that the West wants war, not peace. And not just the power elite, even the people. I am pessimistic, at least in the short and middle term. I try to write, to speak, but it is like trying to stem the ocean tide with my bare hands…”

One of the founders of modern deep learning and the most cited AI scientist Yoshua Bengio responded in the same tune but maybe a bit more optimistically, echoing the approach of Bertrand Russel: “I agree that the world should get rid of nuclear weapons, but I don’t see an easy path except continuing to push for international treaties that gradually build an international consensus, which too slowly creates a kind of world government”.

In summer I added to my project update letter a couple of emerging technologies as samples of not so obvious solutions. One technology “uses neutron induced nuclear resonances’’ and their sensitivity “to the combination of isotopics and geometry” to identify and verify nuclear warheads remotely. [18]

Another, currently under development by the Nobel Prize in Physics Gérard Mourou, will use super-powered lasers to transmute radioactive nuclear isotopes into nonradioactive materials. [19]

The response from Mikhail Shfman was as gloomy as the other responses although with a very tiny spark of hope in the end.

“I do not believe that the problem can be solved through any type of negotiations. At the moment, we see that there can be no trust of any type between “East” and “West”, and there is no way any government involved will make the first step. For Russia, the nuclear weapon is the only way to retain its influence in world affairs.

Even after Putin, decades (many, many decades) must transpire before his successors will show the desire to get rid of nuclear weapons. But then again, North Korea, Pakistan, etc. will remain.

Technology for verification per se is a good first step, but by far not the last”.

In search of the way to overcome the overall apathy, I turned to the heritage of Freeman Dyson. A brilliant mathematical mind who became a physicist, a thinker and a writer, Dyson on a few occasions was involved in resolving problems of potential use of nuclear weapons and nuclear disarmament. In those episodes “Dyson sought to solve political problems with technical tools while withholding his moral judgment”, as a Delhi based defense analyst Abhijnan Rej put it in his tribute to Dyson. [20]

Indeed, a declassified report on military consequences of a US decision to use tactical nuclear weapons in Vietnam prepared by Dyson and his colleagues presents a detailed technical analysis instead of intuitive or moral judgements. [21]

On the other hand, Dyson believed that humankind would need “a worldwide awakening of moral indignation” to achieve nuclear abolition. In early 1980ies he shared his ideas about nuclear disarmament in a series of lectures published as an essay under the title “Bombs and Poetry.”

“There are no compelling technical or political reasons why we and the Russians, and even the French and the Chinese too, should not in time succeed in negotiating our nuclear weapons all the way down to zero. The obstacles are primarily institutional and psychological. Too few of us believe that negotiating down to zero is possible…” Dyson wrote, “The basic issue before us is very simple. Are we, or are we not, ready to face the uncertainties of a world in which nuclear weapons have been negotiated all the way down to zero? If the answer to this question is yes, then there is hope for us and for our grandchildren.” [1]

Although I had a network of scientists in mind I was interested to learn the views of activists of the movement to abolish nuclear weapons.

In 2017 Beatrice Fihn received a Nobel Peace Prize for her efforts which led to the adoption of the Treaty on the Prohibition of Nuclear Weapons.

“The first days after Russia launched its invasion of Ukraine, I have to admit, I felt so much despair and sadness. And it felt like, Ugh, all of this work that we’re doing, not just on the nuclear weapons issue, but on all of the international issues; all of these human rights, international law, it’s all for nothing. They just ignore it, and they just violate it. Why are we even bothering?” she said in a Freakonomics podcast. [22]

“If Russia is not bluffing, can you start forming an anti-nuclear (threat and use) coalition of states? Unlike the anti-Russian coalition, the anti-nuclear coalition will not take the side of Ukraine. It will take a side against breaking nuclear taboos,” I asked her on Twitter.

“The Treaty on the Prohibition of Nuclear Weapons (TPNW) bans threats and use (and possession). The states parties to this treaty (68 states) condemned all nuclear threats in June, and reiterated this in August. It’s a good tool to mobilize governments against these threats,” she responded.

On September 30, 2022, on the day Putin proclaimed the annexation of Ukrainian land, Fihn shared an opinion of another nuclear abolition activist Stephen Young published by Politico.

“For the first time in the nuclear era, one country used loudly issued nuclear threats — repeated just last week — to deter other countries from intervening in a large-scale conventional war of aggression. We have entered the age of “predatory nuclear-weapon states,” he wrote. [23]

Francesca Giovannini is the executive director of the Project on Managing the Atom at the Belfer Center of the Harvard Kennedy School. Young borrowed the term “predatory state” from her article on the site of the Arms Control Association.

After exploring several scenarios of the political fallout of the war in Ukraine, Giovannini finally arrived at the most frightening scenario for her and other activists, “A more likely, less discussed, but equally consequential scenario is that the war in Ukraine will deepen alienation, grievances, and mistrust among non-nuclear-weapon states. Such estrangement could lead to even more fierce political obstructionism against any new nuclear policy and a dangerous institutional paralysis across multiple institutions”. [24]

Daryl G. Kimball, the executive director of the Arms Control Association expressed some interest in my initiative although he wasn’t clear about its concept.

“If Russia or any state uses nuclear weapons, UN member states should pursue a ‘uniting for peace’ resolution to overcome crippling gridlock in the UN Security Council and authorize effective, collective measures to restore the peace and hold Putin accountable under Chapter VII of the UN Charter.

Putin may be cold, calculating, and cruel; but even by his own logic, nuclear weapons use runs counter to his interests, the interest of his regime, and of course the world. Nuclear war made no sense in 1962; it makes no sense in 2022, or ever. It is imperative for responsible leaders to reinforce the nuclear weapons taboo now.” [25]

Nuclear abolition activists and scientists alike see only an international political way towards nuclear disarmament and evaluate perspectives of any progress on this way as rather grim. However, the story of the atomic bomb began as a proposal of a purely scientific and technological solution to a political problem.

Albert Einstein and his friend, a Hungarian physicist Leo Szillard were both pacifists, but their fear of Hitler getting the atomic bomb resulted in their proposal to American government to start a program to develop its own bomb as a deterrent.

An American physicist Robert Oppenheimer was the scientific head of the Manhattan Project that evolved from Einstein’s and Scillard’s initiative. A scientist secured the continuation of the development of the bomb after Hitler was defeated and the original rationale of the bomb as a deterrent became obsolete. Freeman Dyson also describes in his lectures an episode when Oppenheimer traveled to Paris in November 1951 to persuade General Eisenhower — who was then in command of American forces in Europe — that tactical nuclear weapons would help his armies to carry out their mission of defense.

Freeman Dyson asked Oppenheimer, “long after he had lost his security clearance, whether he regretted having fought so hard for tactical nuclear weapons. He said, “No. But to understand what I did then, you would have to see the Air Force war plan as it existed in 1951. That was the Goddamnedest thing I ever saw. Anything, even the war plans we have now, is better than that.” [1]

Sankichi Toge survived the atomic bombing of Hiroshima. His poem “August 6” was published in 1951 in the collection “Poems of the Atomic Bomb”. It was translated into English in 2011 by Karen Thornber, a recipient of the Sibley Prize from the University of Chicago for her translation of the collection of Toge’s poems.

“can we forget that flash?

suddenly 30,000 in the streets disappeared

in the crushed depths of darkness

the shrieks of 50,000 died out

when the swirling yellow smoke thinned

buildings split, bridges collapsed

packed trains rested singed

and a shoreless accumulation of rubble and embers — Hiroshima

before long, a line of naked bodies walking in groups, crying

with skin hanging down like rags

hands on chests

stamping on crumbled brain matter

burnt clothing covering hips

corpses lie on the parade ground like stone images of Jizo, dispersed in all

directions

on the banks of the river, lying one on top of another, a group that had crawled to

a tethered raft

also gradually transformed into corpses beneath the sun’s scorching rays

and in the light of the flames that pierced the evening sky

the place where mother and younger brother were pinned under alive

also was engulfed in flames

and when the morning sun shone on a group of high-school girls

who had fled and were lying

on the floor of the armory, in excrement

their bellies swollen, one eye crushed, half their bodies raw flesh with skin ripped

off, hairless, impossible to tell who was who

all had stopped moving

in a stagnant, offensive smell

the only sound the wings of flies buzzing around metal basins

city of 300,000

can we forget that silence?

in that stillness

the powerful appeal

of the white eye sockets of the wives and children who did not return home

that tore apart our hearts

can it be forgotten?!” [26]

Back in 1939, two days after drafting the letter about the bomb to Franklin Delano Roosevelt, Szilard wrote out his own version of the Ten Commandments. One was, “Do not destroy what you cannot create.” This has an eerie resonance with Richard Feynman’s observation (purportedly on his blackboard at the time of his death): “What I cannot create I do not understand”.

In short, “Do not destroy what you do not understand”. This must surely be the mantra of any curious creature — or institution.

“Those who fight for freedom with the technologies of death end by living in fear of their own technology,” Dyson writes in summary of one of his lectures. Given the long-term consequences of the creation of nuclear weapons, I feel a more apt version of Szilard’s commandment would be, “Do not create what you cannot destroy.” [1]

Many of us experience similar feelings now. “Everyone seems to have so much to say about everything that’s happening right now, but I just feel like I’m both completely out of words and simultaneously just want to scream about how the “nuclear deterrence creates peace and stability” brainwashing got us here,” wrote Beatrice Fihn on Twitter recently. “And never stop working for a better future”, she ended up in spite of all disappointments.

“Sagacity is the essential virtue for the hero of a comedy. It is the art of making the best of a bad job, the art of finding the practical rather than the ideal solution to a problem, the art of lucking out when things look hopeless,” Freeman Dyson concluded in his Bombs and Poetry, “The business of an explorer is not tragedy but survival”. [1]

Mind the Bomb

In May 1945 Leo Szilard took another letter addressed by Einstein to the President, now to the Truman White House. James F. Byrnes, who was about to be named Secretary of State, met with Szillard and dismissed Szilard’s concerns about the forthcoming nuclear arms race with the Soviet Union as well as his proposal about a demonstrational explosion of the A-bomb. Burnes told Szilard that using the A-bomb on Japanese cities might make the Russians “more manageable” after the war. [27]

When Einstein learned about the bombing of Hiroshima he said, “Ach! The world is not ready for it.” [28]

“When I visited Einstein at Princeton shortly after the bomb was dropped at Hiroshima and Nagasaki,” Szilard later recalled, “our conversation turned back six years to the visit on Long Island when we discussed the letter he might write to the President. ‘You see now,’ Einstein said to me, ‘that the ancient Chinese were right. It is not possible to foresee the results of what you do. The only wise thing to do is to take no action — to take absolutely no action.’” [28]

“Aren’t you afraid that we can create something even more dangerous to humankind than nuclear weapons in our effort to eradicate them?” Vitaly Vanchurin asked me in one of our early discussions of the initiative. “We have to take care from the very beginning that it won’t happen,” I responded, taking Vanchurin’s concern rather lightly as a very remote perspective.

In the meantime, this fall a new paper [ Friston et al., 29] provided a solid experimental proof of both the free energy principle and the active inference theory. Maybe I am exaggerating it, but I see synthetic biological intelligence (SBI) presented in that paper as a basal model and a proof of concept of AGI.

Combining the findings in that paper with different works grounded in the free energy principle [Levin et al., 13–17], I had a vision of the emergence of a sentient life form that, unlike currently available dumb and rigid AI systems, will be smart and agile. It will learn from scratch without supervision or reinforcement. It’ll get itself embodied in various substrates and embedded in different environments. It will be able to change its morphology, regenerate, self-replicate and multiply.

What if such a life form would end up developed as a weapon in a military lab or would sooner or later escape a lab and get wild and dangerous? Can we envisage the results? Maybe, we should take no action as Einstein suggested? “Do not create what you cannot destroy.”

A Tiny Spark of Hope

“I don’t think AI can overcome human frailty..”

This FWIW remark was made by Esther Dyson, a brilliant innovative mind in venture capital and technology and the daughter of Freeman Dyson, in her generally supportive response to my email describing the initiative.

Frailty.. but should we overcome frailty? Erwin Schrodinger once predicted that the human species could leapfrog the suicidal evolutionary stage of national egoism and achieve lasting peace between peoples; not due to human virtues but because of human frailty, namely, cowardice. Human species can unite, leveraging this egoistic trait that isn’t yet totally suppressed by the trait of nationalism, altruistic for an individual, but egoistic for a nation, because it demands ultimate individual bravery to sacrifice everything, even life for the sake of nation. [29]

Being stuck at the national egoism stage of evolution means that wars will always be fought. Ants survive at the national egoism stage for millions of years, because their ways of fighting wars don’t evolve. If we humans follow their route and yet keep developing more and more lethal and destructive weapons, we will end up eradicating ourselves.

An individual can thrive without sacrificing itself for the sake of the entire human race, but not any part thereof. Ashby’s law of requisite variety is pressing species to maximize individual deviations to always be ready for new and unexpected challenges of the environment. [30] However, cowardice alone isn’t enough to unite humankind. Atop of it we need intelligence, but..

A peace lover says that we have to give the dictator what he wants to stop the war and prevent its escalation into nuclear Armageddon. Another peace lover claims that, on the contrary, we should fight the dictator till his fall because if he gets what he wants now he will raise stakes and other dictators will follow his example in the future. Such a development will make nuclear Armageddon much more probable.

We can’t tell who is right because in both cases the probability of nuclear Armageddon can’t be excluded. Even if we believe that in one case the probability is much higher it doesn’t mean that the lower probability variant can’t materialize.

Eradicating nukes may satisfy both kinds of peace lovers, but they don’t believe that it’s possible and consider pursuing it a waste of time. Isn’t it a stalemate?

Human intelligence in its current shape and cowardice also aren’t enough. I wonder if the free energy principle, the universal theory of learning, and all the frameworks based on them can be leveraged to incite Schrödinger’s leap — by converging many different minds, natural and artificial? The eradication of nukes might become the first use case of this new level of intelligence.

Shouldn’t we at least try to leverage technologies of life to free us from the fear of technologies of death? Or should we take no action? Cowardice or daring?

Epilogue

After sending a draft of this White Paper to some select colleagues, I received the following from Karl Friston

“This was a riveting read. It made me think seriously about how to apply the free energy principle to self organisation — or perhaps self evidencing [31] — in this setting. Two themes occurred to me. One picks up on Mike [Levin] and colleagues’ work on scale-free, basal cognition that rests heavily upon the notion of individuation — in my parlance, Markov blankets. The key intuition here is that the existence of any ‘thing’ — at any scale — rests upon the autopoiesis or self-assembly of Markov blankets that individuate the constituents of an ensemble (from the cells of an organ through to the institutions of a nation). In other words, to exist as some ‘thing’ is to individuate yourself from every ‘thing’ else. Any ‘thing’ that threatens the integrity of our Markov blankets is an existential threat. For example, if the action of one person can influence too many others by incinerating them (or influencing millions on Twitter). I suspect that this is the pathology that needs tackling; much in the same way that at a cellular level one needs to restore the local integrity of sparse intracellular signalling to prevent cancer. This reminds me of Mike’s notion that cancer — when read in terms of basal cognition — is a delusion; namely a failure to infer the state of affairs hidden beyond one’s Markov blanket.

The second theme relates to the Bayesian mechanics of active inference [32]. You have rightly made much of curiosity as an existential imperative. Technically, this inherits from minimising expected free energy (of the Feynman kind). Expected free energy can be decomposed into expected information gain and expected cost. Expected cost here is simply the violation of prior preferences; namely, the probability of occupying states that are characteristic of me. This decomposition leads to an interesting dialectic: one can minimise expected free energy by resolving uncertainty through curious, information-seeking behaviour or through preference-seeking behaviour to avoid surprising outcomes (e.g., being embarrassed, being hypothermic, or being incinerated). I remember reflecting on this prior to Trump’s election: on the one hand, his success would be surprising given my prior beliefs about America. On the other hand, I was intensely curious to see what would happen if he were elected. I mean this in the sense that the evening news would acquire a special value — an epistemic affordance — to answer the burning question “what has he done today?” If this formulation has any currency, it speaks to increasing the precision of prior preferences to avoid a state of nuclear war while, at the same time, reducing the expected information gain that would ensue from witnessing a nuclear exchange.

I guess one way to do this is to redeploy the film industry’s efforts and expertise — away from fantasy and comic–book entertainment — towards a franchise of blockbusters showing people what it would be like to experience a nuclear war. I thought “Don’t look up” was particularly effective in this regard, much in the spirit of ‘Dr Strangelove’ and ‘Oh what a lovely War’. Interestingly, these classics were probably the source of my teenage nightmares.”

On reading his email, I wondered whether curiosity about the consequences of a nuclear exchange may be a testament to the old adage “curiosity killed the cat”.

Checking the origins of that adage I discovered that it emerged only in the end of the Nineteenth century. The original, much older English proverb is saying “care killed a cat.” William Shakespeare was among those who used it. Circa 1599 in his comedy Much Ado About Nothing he wrote, “What, courage man! what though care killed a cat, thou hast mettle enough in thee to kill care.” [33]

However, as we now know from Thorndike’s experiments and Friston’s theory, it was not just any courage but specifically daring induced by curiosity that killed carefulness in cats thus helping them to find solutions to Thorndike’s puzzles and — ways out of the box. I wonder if we humans can overcome nuclear and other existential threats by following the example of Thorndike’s cats?

References

1. Dyson, Freeman. Bombs and Poetry. The Tanner Lectures On Human Values Delivered at Brasenose College, Oxford University May 5,12, and 19, 1982

2. Friston, K. The free-energy principle: a unified brain theory?. Nat Rev Neurosci 11, 127–138 (2010). https://doi.org/10.1038/nrn2787

3. Chris Fields, Karl Friston, James F. Glazebrook, Michael Levin. A free energy principle for generic quantum systems. Progress in Biophysics and Molecular Biology 173: 36–59, 2022

4. Thomas Parr, Giovanni Pezzulo, Karl J. Friston. Active Inference: The Free Energy Principle in Mind, Brain, and Behavior. DOI: https://doi.org/10.7551/mitpress/12441.001.0001 ISBN (electronic): 9780262369978 Publisher: The MIT Press Published: 2022

5. Berlyne, D.E., Novelty and curiosity as determinants of explanatory behaviour. British Journal of Psychology-General Section, 1950. 41: p. 68–80.

6. Schmidhuber, J., Formal Theory of Creativity, Fun, and Intrinsic Motivation (1990–2010). Ieee Transactions on Autonomous Mental Development, 2010. 2(3): p. 230–247.

7. Still, S. and D. Precup, An information-theoretic approach to curiosity-driven reinforcement learning. Theory Biosci, 2012. 131(3): p. 139–48.

8. Friston, K.J., et al., Active Inference, Curiosity and Insight. Neural Comput, 2017. 29(10): p. 2633–2683.

9. Schwartenbeck, P., et al., Computational mechanisms of curiosity and goal-directed exploration. Elife, 2019. 8: p. e41703.

10. Thorndike, Edward (1898) Animal intelligence: An experimental study of the associative processes in animals. Monograph Supplement №8

11. Pavlov, Ivan (1933) Psychology as a Science. Unpublished and Little-known Materials of I.P. Pavlov, in Russian (1975)

12. Vanchurin V. The World as a Neural Network. Entropy (Basel). 2020 Oct 26;22(11):1210. doi: 10.3390/e22111210. PMID: 33286978; PMCID: PMC7712105.

13. S. Kriegman, D. Blackiston, M. Levin, J. Bongard, A scalable pipeline for designing reconfigurable organisms. Proc. Natl. Acad. Sci. U.S.A. 117, 1853–1859 (2020).

14. S. Kriegman, D. Blackiston, M. Levin, J. Bongard, Kinematic self-replication in reconfigurable organisms. Proc. Natl. Acad. Sci. U.S.A. November 29, 2021. 118 (49) e2112672118. https://doi.org/10.1073/pnas.2112672118

15. Levin M, Pietak AM, Bischof J. Planarian regeneration as a model of anatomical homeostasis: Recent progress in biophysical and computational approaches. Semin Cell Dev Biol. 2019 Mar;87:125–144. doi: 10.1016/j.semcdb.2018.04.003. Epub 2018 May 1. PMID: 29635019; PMCID: PMC6234102.

16. Nirosha J. Murugan , Hannah J. Vigran, Kelsie A. Miller, Annie Golding, Quang L. Pham, Megan M. Sperry, Cody Rasmussen-Ivey, Anna W. Kane, David L. Kaplan, Michael Levin. Acute multidrug delivery via a wearable bioreactor facilitates long-term limb regeneration and functional recovery in adult Xenopus laevis. SCIENCE ADVANCES. 26 Jan 2022. Vol 8, Issue 4. DOI: 10.1126/sciadv.abj2164

17. Levin Michael. Technological Approach to Mind Everywhere: An Experimentally-Grounded Framework for Understanding Diverse Bodies and Minds. Frontiers in Systems Neuroscience,16, 2022. DOI=10.3389/fnsys.2022.768201 URL=https://www.frontiersin.org/articles/10.3389/fnsys.2022.768201.

18.Engel, E.M., Danagoulian, A. A physically cryptographic warhead verification system using neutron induced nuclear resonances. Nat Commun 10, 4433 (2019). https://doi.org/10.1038/s41467-019-12386-0

19. Gérard Mourou, Super-powered lasers to transform nuclear waste, Polytechnique Insights, January 27th, 2021

20. Abhijnan Rej, Freeman Dyson and the Challenges of the First Nuclear Age, Science,: the Wire, 07/03/2020

21. F. J. Dyson, R. Gomer, S. Weinberg, S. C. Wright. Tactical Nuclear Weapons In Southeast Asia. Institute For Defense Analyses, Jason Division, Contract DAHC15 67 C 0011, PublIshed March 1967

22. Steven D. Lewitt, Beatrice Fihn, Bombs Away, Freaconomics, podcast, April15, 2022

23. Stephen Young. The Age of Predatory Nuclear-Weapon States Has Arrived. Politico, September 30, 2022

24. Francesca Giovannini. Negative Security Assurances After Russia’s Invasion of Ukraine. Arms Control Association. July/August 2022

25. Daryl G. Kimball. No Viable ‘Nuclear Option’ for Russia in Ukraine. Arms Control Association. October 2022

26. TŌGE Sankichi. August 6. Hiroshima In Four Poems. PBS

27. William Lanouette. Einstein And Szilard In Princeton. The Lewis B. Cuyler Lecture Historical Society of Princeton, Nassau Club, Princeton, February 9, 2011

28. Brett J. Kagan, Andy C. Kitchen, Nhi T. Tran, Forough Habibollahi, Moein Khajehnejad, Bradyn J. Parker, Anjali Bhat, Ben Rollo, Adeel Razi, Karl J. Friston,

In Vitro Neurons Learn And Exhibit Sentience When Embodied In A Simulated Game-World, Neuron, 2022, ISSN 0896–6273, https://doi.org/10.1016/j.neuron.2022.09.001.

29. Erwin Schrodinger. Mind and Matter. Cambridge University Press.1958.

30. Ashby W.R. (1958) Requisite variety and its implications for the control of complex systems, Cybernetica 1:2, p. 83–99.

31. Hohwy, J., The Self-Evidencing Brain. Nous, 2016. 50(2): p. 259–285.

32. Ramstead, M.J.D., et al. On Bayesian Mechanics: A Physics of and by Beliefs. 2022. arXiv:2205.11543.

33. Shakespeare, William. Much Ado About Nothing. Random House Publishing Group, 2009