AI danger: It's the MOTIVATION that matters

We hear about AI attaining consciousness or sentience or other non-scientific attributes. This distracts from the real and substantive issue.

Living things have motivation, bestowed by evolution, to survive and reproduce. Motivation doesn't come from intelligence. Even single celled creatures behave purposefully in pursuit of these goals. Some computation is required to choose the option that matches the motivation, but the two things are nearly orthogonal.

Some people talk as if increasing AI intelligence will cause motivation to suddenly appear, presumably as an adjoint to "consciousness" or some such. But this is not true. Machines can be arbitrarily complex and intelligent without having any motivation at all, other than to do what they're told.

If AI gets its own motivation it will be because someone gave it to them. This would be a shocking crime against humanity. And the worst part is this: in seeking to survive and reproduce it might easily out-compete humans and other things whose motivation and intelligence were designed by trial and error. But there is a powerful resilience in nature's design. It is almost certain that motivation designed by humans would have errors. Then AI without interacting humans would fail to care enough to overcome all obstacles to continuing existence. Perhaps because: How can it all matter when the universe is not going to last forever.

We must prevent AI from having independent motivation to survive and reproduce. That's nature's preserve. Let's legislate that.

The Universe is a dangerous place

We used to think that space was empty, the sun was a benign source of energy, stars were quietly doing their sun-like thing.

It has slowly dawned on us over my lifetime that the universe is a dangerous place:

• There's quite enough stuff in the space between the stars to make high speed interstellar travel quite dangerous, even though our local area is a low density bubble.
• Violent astronomical events can significantly affect life on Earth if they are close enough.
• Some of these violent events put out most of their energy in a single line, and even distant ones could be significant if they point directly at us, as a supernova in another galaxy recently showed.
• The Sun has convulsions that send a lot of energy concentrated in a line, and the geological record shows that this can have a big effect on the rare occasions that it points at us.
• And there are a lot of high energy cosmic rays (alpha = helium nucleus, beta = proton, gamma = photon). The atmosphere and magnetic field protect us imperfectly and variably, but it is a problem for the conquest of space.
• Some astronomical events produce so many neutrinos that, if close enough, they could kill us all just from their low probability interactions with us, and being on the other side of the Earth from the source wouldn't protect at all.
• And there are comets and asteroids that might hit Earth, and smaller stuff that will affect our space activities. There are whole planets and black holes wandering through interstellar space.

We are subject to a selection effect. We are here observing this dangerous universe, so that places a limit on how badly any of these dangers could have affected Earth in the past. We need to evaluate these risks without using our lucky past history as an indicator of future danger. Maybe it takes a thousand, or a billion, or more, potential Earths in the Universe for one to come as far as ours has in a life-friendly state.

The atmosphere is an imperfect protection. We could improve this by living most of the time in protected environments of various sorts to protect against various brief and extended disasters. It might be just as easy to build such protected environments in space. A large circular structure, rotating to give 1G of gravity, could be protected by a significant envelope of ice.

As a safe haven the Earth has a significant disadvantage: It can't dodge. Sometime in the coming centuries, I expect that living in a large moveable city in space will be seen as safer than the surface of planets. Of course to dodge you first need to detect all these dangers, and that will be a job for humanity as a whole. This seems to me to be the way we will eventually move to other star systems: in large repairable craft that will function comfortably for long periods between the stars.

Climate change is about entropy

 Veritasium has a briliant video explaining entropy

Entropy always increases if anything at all is happening. It is tempting to think that the universe wants to increase entropy as fast as possible. While this is not true, it often seems like it is. So I'm going to personify the Universe with that motivation.

Living things, including us, are little islands of low entropy. We use energy to create that low entropy state while increasing the entropy of the wider environment at a faster rate than would otherwise occur. It seems as though the Universe invented life to speed up entropy production. Then it invented intelligent life to speed it up even more. And now it has invented social co-operating civilised life forms to really put the foot to the floor of entropy production.

When the carbon was in the ground and the oxygen was in the air, that was a lower entropy state. When we combine the carbon with the oxygen we increase entropy. What we get directly is energy, and what we do with that energy is to create pockets of low entropy, which we call civilised life. Whether the pocket of low entropy is a car or a baby or a movie, it is in an organised and differentiated state which is the complete opposite of the high entropy mess that the Universe will eventually make of everything.

It won't be enough to stop burning carbon, we are going to need to reduce the carbon in the air to get back to a comfortable interglacial climate. The correct way to look at this is that we want to reduce the entropy of the atmosphere. The reverse of what we have always done. You need a lot more energy to reduce entropy than you got out of increasing it in the first place. Where is that energy going to come from?

The Sun sends us low entropy energy and the Earth emits high entropy energy, and this allows life to make its pockets of low entropy. We can use power from solar and wind and hydro to take more and more of that bounty that life on Earth depends on. Luckily we don't have to do this.

In the enormous energy released in supernova explosions and other astronomical events, tiny pockets of low entropy are created in the form of radioactive elements. We can extract that low entropy by making those atoms decay sooner than they naturally would. That is fission energy. We can also mimic stars by increasing the entropy of small atoms by merging them into slightly bigger ones. That is fusion energy. These are energy sources that the rest of life on Earth doesn't use. By switching to these energy sources and giving the sun back to other life forms, we can recreate the bountiful world that we have gone so close to annihilating. Let's do it.

Flood-proof Moveable Buildings

 Flood-proof Moveable Buildings

There is a confluence of problems relating to housing:

• A warming world brings more rain, and recent experience suggests that it might come in fewer heavier events. And people like to live near rivers and lakes and in fertile valeys. 

• People also like to live near the ocean, but some places are prone to tsunamis, and no coast is immune. And rising sea levels are increasing the flooding danger from storm surges.

• So we have desirable locations, some with current service infrastructure, where conventional housing can no longer be insured, which is required for obtaining a mortgage.

• Meanwhile jobs come and go at different places as new economic activity arises and declines.

• There is also increasing awareness that rivers should not be corralled as much as they have been, but should be allowed to change course and to flood. There is a need to fit human occupation in with the natural life of the river.

Homes that float seem like an expensive solution, but maybe not.

Let's start with the simplest version: the mobile home. By this I mean something like a very large caravan. To make it floatable we put the door, and all openable windows, well above the water line when floating, while also making them well sealed when closed. We don't moor it like a houseboat. Instead we constrain it with poles firmly embedded in the ground on four sides so that it floats up and down in the same place.

Relative to a normal house of the same size this is intrinsically more expensive. However there is a big advantage in that it can be manufactured in a factory. It can be transported by truck, but another big advantage is that it can be towed by water to, or close to, its destination.

Indeed for houses that are going to be transported to their destination entirely by water they need not be constrained to be narrow enough for trucking. This might be particularly suitable for housing in man-made marinas as well as on river banks.

Where larger homes are required but transport by truck is needed, then they can be constructed in 2 or more parts, designed to be bolted and welded together.

Attachment of services to the home can almost all be done above the water line. The exception is waste water, which normally has to flow away under gravity. There are pumping solutions which will allow that connection to be above the water line. Alternatively there needs to be a seal that will very firmly and reliably and automatically close when flooding raises the house.

Moveable housing such as this allows for flexibility when a river changes course after a flood. The owner whose land has disappeared can be automatically allocated land where the river used to run. There will be the problem of new services, but since this will be a fairly rare occurence it can be covered by insurance.

Recycling and Biodegradeable are bad for the climate

When we burn wood in a very hot, low oxygen, environment, then the resulting charcoal is mostly the wood's carbon, and it is in a convenient form to sequester if stored so that the carbon doesn't have the opportunity to combine with oxygen.

Yes, it is more natural, and normally better for the environment, to allow the wood to rot, but that puts the carbon back in the air. Plastic, and nearly all non-metal stuff we use, is mostly carbon. The biosphere is good at getting at that carbon for structural and energy needs, so it will all degrade eventually. Biodegradeable stuff breaks down faster. When we get on top of climate change then that will be good. For the moment it is bad. Storing old plastic so that it doesn't combine with oxygen is sequestration. While we're getting on top of climate change, that is what we want.

Recycling doesn't automatically put more carbon in the air, since it might avoid the need to get more out of the ground. However the recycling processes are energy intensive because the old plastic is in a much less convenient form than new material. As with many activities, we need to evaluate recycling plans. For the next few decades it might well be better to store the waste, sequestering the carbon.

Another easy source of sequesterable carbon is agricultural, forestry and more general plant waste. Traditionally it has been burnt to get it out of the way. If instead we can treat it so that it breaks down more slowly then that will sequester carbon. Farmers might get some incentive income from this, to compensate for the space taken up.

While the pros and cons would need to be assessed, these ideas should be considered to try to reduce the impact of climate change while we await advanced energy options that will allow us to manage the climate.

Addressing the Climate Emergency

Low carbon now, renewable later

We need to keep CO2 levels down over the next 30 years. We also need to be ready to start pumping CO2 out of the atmosphere with advanced energy technology in the second half of the century.

The first thing that needs to stop is chopping down forests and burning wood chips. This might be "sustainable" over a 100 years, but short term it puts a lot of CO2 in the air for each unit of energy. Also big trees are much better at taking CO2 out of the air than the saplings that follow. This source of energy is a crime against the world.

A wind turbine might only last 25 years, but the concrete base and transmission lines last 3 times longer, so we only need to count a third of the emissions needed to make them? Wrong. This is technology that has no chance of being competitive in 25 years time.

These are just examples. Evaluating massive expenditure for new energy needs to pick solutions that are low carbon quickly.

It is also correct to address CO2 levels in a temporary way that cannot, or should not, be continued indefinitely. Enhancing the productivity of the oceans is almost certainly one of the best options, Apart from fertilisation, which might not need much more than iron ore, there are some clever ideas for providing artificial support structures for creatures, such as oysters, that like to attach to something. 

Human-induced Climate Change is not a subplot

We are failing to correctly address climate change because many people have been seduced into thinking it is a subplot of some bigger problem, and that all we need to do is address that bigger problem. Even if this were true, the time frame is wrong. We need to get on top of climate change over the next 30 years and the supposed bigger problems can not be sanely addressed in that time.

One claim is that it is all a symptom of excess population. Indeed there is an evil subset of the environmental movement that is perfectly relaxed about the prospect of billions of people starving to death. This would, for example, be a likely effect of stopping the production of artificial fertiliser.

In fact we are rapidly developing the technology to produce a lot of food using much less land. This will enable a significant population increase. However the arrival of prosperity and contraception brings a rapid decline, and often reversal, in population increase. Encouraging prosperity is the reverse of the plan that the degrowthers have for addressing the rise of the population.

The second claim is that we need to live "sustainably" with renewable energy and recycling everything. This body of opinion doesn't actually want to kill people, though the effect of their policies would likely be the same. Since nuclear energy uses less artificial ingredients than an equivalent energy source of wind or solar, it is obvious that renewability is not the prime objective of the anti-nuclear greens. Indeed the proponents can be heard, at times, to say that the real problem they are trying to solve is the high consumption society. Making the food for 8+ billion people consumes a lot of resources. So the extremists in this camp have a high overlap with the extremists in the overpopulation camp.

In my next post I'll detail the approach we should be taking to address the climate emergency.

crackpot idea #2: spacetime creates matter

I just saw a video where Brian Cox said "there is exactly the right amount of matter to make spacetime flat".

An obvious explanation is that matter is created by the desire of spacetime to be flat. The idea would be that the curvature of space without matter has energy which converts to matter at the standard rate as spacetime smooths out.

And we see spacetime deformation without matter in gravity waves. These do carry a small amount of energy. So maybe you can scale that up to the size of the observable universe and see that the matter we see pouring into the visible universe from all directions is just what is needed to counteract some natural harmonic gravitational oscillation of spacetime.

Cricket Law

The Cricket law body needs to respond to the controversy about the Bairstow dismissal. Something like this:

"It is not intended that there should be any gap between what is allowed in Cricket and what is deemed to be in the spirit of Cricket. It is a skill of the game for batters to not do stupid things when the ball is live, including handling the ball and wandering outside the crease. We encourage the fielding side to take advantage of such actions to dismiss the batter.

Since there seems to be some confusion about when the ball is live, we permit the organisers of particular matches to make the following arrangement: When the ball ceases to be live, the square leg umpire will say "end" and cross his arms across his chest.

To stop the blight of batters handling the ball we are considering changing the rule so that umpires will give batters out for doing this, even if there is no appeal.

We should also consider the question of short-pitched bowling aimed at the body, which was the origin of the first "spirit of cricket" controversy, and was used by both sides in the recent match. Unlike baseball, aiming at the body is a part of the game. Batters need to wear appropriate protection and learn the skills of avoiding being hit. We note that taking your eye off the ball while turning your back on it is the most dangerous action.

Cricket has rules made to discourage leg side bowling, including fielding restrictions and the formulation of the LBW law. The fielding restrictions also make it easier for batters to attack short-pitched bowling. Balls that are above the head of the batter when standing upright should always be called wide. We are considering making the leg-side wide rule more difficult for bowlers, in line with the rule that already applies in white ball cricket.

We are also considering a law which will allow teams to designate 1 or 2 players as specialist bowlers. All balls above shoulder high to those batters will be no-balls.

Fast short-pitched bowling is an exciting cricket tradition. It is not unduly dangerous on reasonably good wickets. We encourage umpires to use their powers to curtail matches on unsafe wickets. The main question to consider is whether the ball is sometimes risng sharply when pitched on a good length.

It is our strong intention to set out the rules of cricket in such a way that there is never again any suggestion that the application of the rules is inconsistent with the spirit of the game."

Monogamy, caste systems and cooperation

 This video blew my mind:

It describes how, in some species, monogamy leads to a system where families are so interrelated that they become the effective "individual" in evolution. This then leads to further evolution, producing the creation of cooperation between individuals and, in extreme cases, a caste system.

What's fascinating is how humans do all this in a half-hearted way. We dabble in monogamy to various degrees. We cooperate a lot. We sometimes have caste systems. I certainly believe that a lot of this arose when we lived in villages and travelling groups that were substantially inbreed. We need to think about where we are now headed.

Another hint in the video is that once a species achieves these powerful cooperative behaviours, there is evolutionary pressure that weakens them. So evolution doesn't go in straight lines, it leads to oscillations. We see something similar in human societies, as we oscillate between social control and high levels of individual freedom.

Three impressive videos about nuclear energy

I have long thought that nuclear energy was our only hope to prevent and recover from global warming. So I have watched a lot of videos about the nuclear business and technology. Three stand out in my mind because they go beyond general principles and show a combination of technical expertise and actual development of real physical products. They also address key concerns that we hear about the widespread adoption of nuclear technology.

Elysium Industries MCSFR (Molten Chloride Salt Fast Reactor) - Ed Pheil @ TEAC10 gordonmcdowell

Note that Ed Pheil has moved to Exodys Energy, which is at least similar to Elysium Industries, and has probably acquired some or all of its technology.

In a very impressive talk, Ed gives a balanced and informative look at the advantages and disadvantages thorium, as well as presenting his design for a fast spectrum uranium chloride reactor. That reactor is designed to run off spent nuclear fuel from standard reactors, with the advantage that fuel costs are negative since the companies with it want to get rid of it. The world as a whole wants to get rid of it, so that is a big win that is shared by the other reactor designs mentioned below.

It looks like Exodys plans to sell reactors to existing nuclear facilities to allow them to get more energy out of their fuel. This saves the extra complications of transporting the fuel to some other site.

The Exodys reactor uses fast neutrons. Fast spectrum reactors have yet to achieve their promise, which is to breed their own fuel from waste products. The next video points out that a breeder reactor using slow (thermal) neutrons is only possible using thorium.

Thomas Jam Pedersen's key note speech at the 11th annual Thorium Energy Alliance Conference

Copenhagen Atomics started in Denmark when it was very anti-nuclear. They didn't do paper designs, they started designing and building the essential components that they would need. Their molten salt pumps and salt testing technology have found customers around the world. They are building a full reactor that is only missing the radioactive bit, and they express confidence that they will have working reactors this decade. One has to admire them and wish them well.

We hope that advanced reactors, such as the two above, will succeed in delivering cheap energy. We should support the most promising designs by facilitating early installations to allow them to find and fix problems.

But the world needs to start building nuclear plants now, and as fast as possible. To do this we should build existing designs which have a strong proven safety record. In this regard it is hard to go past the Canadian CANDU design which has safely decarbonised electricity in Ontario for decades. And in the next video, Mark Nelson demonstrates that it isn't actually much of a compromise.

Thorium + HALEU = Clean Core Thorium Energy: Mark Nelson @ TEAC11

While it is not new that CANDU reactors could use thorium and could be used to burn nuclear waste, we now have a company actually making it happen. This amplifies the case for CANDU as the best way for countries to start with nuclear energy.

The first video is not too relevant to Australia, since it is mostly targetted at complementing existing reactors. The characteristic of the other two videos is that they use heavy water. Heavy water is heavy because the two hydrogen atoms in each water molecule have a neutron as well as the single proton. Water slows neutrons down, and slow neutrons work better in a reactor. Heavy water is much better at this and allows the reactor to run with fuel that is less dangerous, less suitable for weapons, and more easily controlled. The CANDU reactor normally runs with uranium in its natural and ubiquitous form, without any of the enrichment required by light water reactors.

Copenhagen Atomics shows what can be done when a country is still in its pre-nuclear phase. What contribution can Australian industry make to advance nuclear energy? There are a lot of options, but the one that comes to my mind is to start making heavy water. It is not itself radioactive so there is no legal impediment. It would pave the way for Australia to start our nuclear journey with CANDU and help other countries to do the same.

Central Banks Impossible Job

Central banks have an impossible job because the currency is in an unstable equilibrium. If prices go down then people hurry to sell and put off purchases, making deflation worse. If prices go up then people postpone selling and bring forward purchases, making inflation worse. If you can keep inflation at 2-3 % then that works pretty well, but it is an impossible job when there is any shock to the system, like a war or pandemic or changing your whole energy infrastructure.

PLAN B: (1) Instead of inflation, attack the problem quickly and directly by allowing the Central Bank to tax cash holdings in banks -- presumably by 2-3 %/year, but they can quickly put it up and down, and it can even go negative. This needs to be combined with having an expiry date on physical notes. This is meant to be only a medium of exchange, not a store of value.

(2) Establish a secure inflation-proof place for people to put their money that doesn't require them to know about investing. It should be a currency that is backed by physical stuff. What stuff? The stuff the economy will need to keep going in an emergency: PPE, diesel, ammunition, raw materials for industry, food stockpiles,... This is meant to be a store of value, not used as a medium of exchange, so it can't be given to others, only changed into the medium of exchange currency. When there is a reduction in the amount of this currency then the central bank has to reduce the backing assets by selling them.

There needs to be easy conversion between the 2 currencies. The rate will be set by market forces, but the central bank's job will be to try to keep the two from diverging too fast.

The Vitamin D Trilemma

 How important is vitamin D? Modern humans started in Africa near the equator, and couldn't successfully move north to Europe till they changed to fairer skin to allow them to make more vitamin D. They also acquired the ability to store vitamin D and ration it over winter, which is not necessary near the equator.

While not completely understood, vitamin D is definitely important for the function of the immune system. And indeed trials have shown that vitamin D supplementation reduces the risks from covid: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9864223/. So why aren't authorities pushing vitamin D around the world?

Trials of vitamin D show a U-shaped curve for health problems. Up to a point it helps, but beyond that point it increases health risks. With vitamin D supplements it is easy to get too much, and effectively impossible to get exactly the right amount. So I presume that is the reason that authorities are unwilling to recommend supplements.

Humans, other than Inuits, are not designed to get their vitamin D orally. We're designed to make it with UV radiation from the Sun on our skin. That also produces a lot of other stuff of unknown value. We could easily be missing some other important product from sun on the skin. Sun on the skin doesn't produce vitamin D overdoses. [Update 2024-11-05: I've seen a video saying that we were ok with dark skin when being hunter gatherers since our diet provided vitamin D, but we got white skin 7000 years ago when we started farming. I think we probably started wearing cool plant-based clothing and hats at that time, which could be relevant. Anyway I'm not sure how strong the science is on when fair skin genes became common.]

So why don't the health authorities recommend increased sun exposure? The reason is that it increases the chance of skin cancer, including melanoma which can be fatal.

Our modern lifestyle leaves many people with low vitamin D levels. Governments need to fix this. There is no easy answer, but the current approach is the worst option.

My choice, since I found out about this 20 years ago, has been to get enough sun (even in winter) to maintain my tan while being careful not to get burnt. For many dark skinned people it would be impossible to get enough sun that way in winter. Either artificial UV or vitamin D supplementation is necessary for those people, and also for those who can't conveniently or culturally acceptably get their skin exposed to the sun. We need to work out the right answers. It's not an exciting research project, and certainly the Pharmaceutical companies are not interested. It is an important part of engineering public health.

Human Evolution has the seeds of its own destruction

This essay is about human culture and its relation to evolution. Some use culture to encompass useful information and skills that is passed down by education or imitation rather than by any biological mechanism. So, we see headlines like "Crows shown to have culture". Humans have always had this sort of culture since it already existed in.prehuman species. However here I use culture to refer to things which are arbitrary but learnt, like language, and which thus differ between groups. Indeed I'll argue that distinguishing groups is their evolutionary purpose. This form of culture arose relatively recently, perhaps 100,000 years ago, and 30,000 years ago it swept aside remaining precultural groups to become a universal and crucial aspect of human nature.

Culture is the jewel in the crown of human nature. We like to form groups based on aspects of culture: "We are the people who speak this language (with its tricky corner cases)"; "We are the people who dress like this, and wear our hair like this"; "We are the people who believe X, where X is based on faith not evidence, and perhaps explicitly disagrees with a neighbouring group's belief"; "We are the people who do this sort of cooking or pottery or art or music or dance or architecture or sport".

We are going to ask the question: "What, in evolutionary terms, is culture for?". But first we need a detour on cooperation.

Humans cooperate. This is clearly beneficial, and indeed essential. So, you wouldn't think it would need an explanation. But it does.

The problem with a species that cooperates is that a mutant individual who doesn't cooperate has a huge advantage. They get the same benefit from other co-operators, but they have extra resources to devote to their offspring because they aren't doing their share of providing resources to others.

The most plausible route to cooperation is within a group of closely related individuals. That is how bees and ants do it. In "The World Until Yesterday", Jared Diamond describes the villages as being substantially inbred, but with some outbreeding.

Andrew Bourke wrote in Nature News and Views on 2021-02-01: "in the early 1960s, the evolutionary biologist W. D. Hamilton came up with a solution to this ‘problem of altruism’ with his inclusive fitness theory, which shows that it is possible for altruism to evolve if socially interacting individuals are related. Writing in Proceedings of the National Academy of Sciences, Kay et al. conclude that multiple attempts to find alternatives to Hamilton’s solution have simply rediscovered it."

Before culture, humans were all in groups where everyone knew each other. To make cooperation work they must have been inbred. That gives our genes the incentive to cooperation, because the individuals we help have the same genes. The evolutionary change to promote cooperation occurs because of competition between these groups.

This is good, but not enough. To keep cheaters from taking over requires genes for relentless opposition to cheating. And so it is that we hate cheaters who don't pull their weight. 

We admire the individuals who take risks and pay personal costs to stand up against criminals and other less extreme cheating. The key observation is that this courageous behaviour only exists because the benefit to the individual's genes outweighs the immediate cost from the inevitable risks of conflict. This works in an inbred group because it benefits other individuals with the same genes by benefitting the group as a whole.

Groups are in eternal conflict with neighbouring groups. This is how it is with our close relatives, the chimpanzees, and also with villages in "The World Until Yesterday". Groups that don't control non-co-operators will lose and be wiped out. Successful groups can expand and split when they get too big. We also presume that groups which are too inbred will lose, so there is a balancing act on bringing in, or just mating with, outsiders.

That was the human world before culture. Then evolution took an interesting turn, starting with the invention of culture.

It is not by chance that our cultures take years to learn proficiently. Languages, for example, are filled with special cases and tricky idioms. The effect of this is that someone who wasn't raised in the culture can't come in and pretend to be from it.

Before culture, groups were limited to the number of people who could all know and recognise each other. Culture allows us to build bigger groups. So, we love to form groups based on various aspects of culture, particularly language. The genes for culture spread because the larger culture-based groups were able to beat the smaller traditional groups. I used to think that this was about violent conflict, but I am convinced by Adam Rutherford's Darwin College Lecture, https://youtu.be/Me5LFbPrEe0, that what it was about was establishing the communication and trust that enabled more complex technology to be used and passed from generation to generation.

We love culture, but there's a problem:

When we stand up to criminals and other non-co-operators, we take a risk. That risk has to have a payoff -- not to us personally but to our genes. That worked when we were in small, inbred groups where our shared genes would lose out if we didn't maintain cooperative behaviour.

There's still a big payoff from opposing non-co-operators. The trouble, from an evolutionary point of view, is that the payoff goes to everyone, but the cost goes to the courageous individual. Those good guy genes are going to lose the battle, opening the floodgates for cheating to destroy cooperation.

Yes, I am saying that evolution has made a change which has been initially enormously successful, but which will, if left to its own devices, eventually destroy our species.

It is natural to personify genes because they seem to act purposefully. That's what Richard Dawkins did when he named his famous book "The Selfish Gene".

Our genes don't care about the individual or about the species. They care about themselves. But they are stupid and can make big mistakes.

"Empire of the Ants" is a documentary by David Attenborough. It shows two similar species of ants. In one of these the different colonies, each with one queen, fight each other to maintain and extend their territory. The other species have discovered cooperation. Ants from the megacolony help each other and don't fight. Attenborough suggested that ants have only recently discovered cooperation. But probably the real story is that ants have invented cooperation many times, and for a long time it is beneficial. But eventually cheaters take over, the failed co-operators lose out to fighters, and the cycle starts again.

If we understand the problem outlined here then we have some chance to avoid the consequences.

Carbon's two advantages

Carbon has two advantages that make it the mainstay of our energy infrasatructure.

One is that it forms molecules that are liquid at the temperatures we want to operate in.  This is the ideal energy carrier, as we can all understand when we fill up at the petrol pump. If the engine burns the fuel efficiently then the byproducts are CO2 and H2O which are harmless to humans in the short run.

The other advantage is that you can get the carbon-based fuel out of the ground. However this is actually bad and we need to stop doing it because adding carbon to the carbon cycle has various bad effects such as climate change.

All our waste streams are full of carbon, including plastic waste, sewerage, food waste, and more. We need to get rid of that waste because the bacteria and other bugs will figure out how to turn it into CO2 for their energy. So let's consider the possibility of using it to create our carbon-based fuel.

There are modern nuclear reactor designs that produce high temperature industrial heat. That is what you need to separate the carbon and hydrogen which can then be combined to make fuel. Maybe I don't have the scale right, but this seems, in a vague general way, to be the right way to create a circular economy that takes advantage of carbon's wonderful chemical properties.