Do you think your AI research has implications for this situation? It seems to me that going from our idiot god, toward self-engineering intelligence is a step up by an order of magnitude, so that such a "metaengineer" could, in fact, choose to optimize for species survival, or some other virtue that it chose.
I think the notion of "species" for a superintelligence doesn't really follow because I don't see the idea of "individual" surviving unambiguously in such a scenario, but I think my question still makes some sense: if evolution kills its creations by selecting for short term individual fitness at the expense of the species, do you think the next step of life, having been intelligently designed, will change the nature of that problem entirely?
"If the human species was going to stay biological for another century, it would be a good idea to start cloning Gandhi."
I would bet against that being a good idea. It's not at all obvious to me that Gandhi was either instrumentally beneficial for either mankind or India, or that he took actions that could reasonably have been expected to be beneficial, but even assuming that he did it's not at all obvious that "great men", especially political "great men" on average make the world a better rather than a worse place. If they on average make the world worse, then you are betting on genetic variation specifying not only "be a politically great man" but also "be one of the rare exceptions within that category who actually do good things". A bunch of clones of Gandhi might have effects similar to a bunch of clones of Washington, Napoleon, Hitler, Alexander, Lincoln, Caesar, etc, which probably leaves us worse off.
The problems being discussed arise not because of biology, nor because biological organisms are not directly designed. The issue is replication with variation and the necessary historical consequences of this. The only way to avoid the consequences would be to stop replicating or to stop the variation.
The variation cannot be stopped, since no two physical things will ever be exactly alike. And if the replication is stopped, the whole system will go extinct, since no particular individual physical object will last forever. There is therefore no way to avoid the consequences of evolution: they are not biological consequences, but consequences of the laws of physics and logic. There is no way around them.
Unknown wrote: The variation cannot be stopped, since no two physical things will ever be exactly alike.
Assuming we can neglect position in space and movement vectors, different individual elemental particles (e.g. electrons) are already exactly alike. Individual atoms are frequently exactly alike, and the same can be said for molecules. We don't have a way to precisely copy macroscopic objects yet, but that might change with further technological development. Do you think that Molecular Nanotechnology is fundamentally impossible? If so, why?
And even if you can't build machines with absolute precision, that doesn't mean you can't build them in a way that will prevent them from being actively destructive. If you're worried that machines you build may become destructive because of construction errors, add a test stage to the construction process, and/or self-test components that will cause the machine to shut down harmlessly when an error is detected. Noticing random corruption isn't very hard, from a computer science perspective. There are plenty of cheap and effective hash functions available. By adding a bit of redundant data, you can drive the probability of a random corruption being unnotived down to infinitesimal levels.
Unknown, you're making the assumption that the entity or entities in question will continue to replicate in a fundamentally similar manner to biological organisms, and I think that's a flawed assumption. My person bias is toward believing that an AI would not so much replicate, as envelope. Even if I am wrong, Eliezer's previous points about our concept of intelligence being a very small portion of the space of possible intelligences holds here.
Sebastian, rather than argue with that, I can point out that even if the variation can be stopped in principle, at least to such a degree that there will be no significant variation within the lifetime of the universe, there is no reason to think that the variation will actually be stopped, unless the human race goes extinct first. For the presently existing variation in the human race ensures the existence of deliberate efforts to produce more variation, and so the amount of variation can be expected to increase, not decrease or vanish.
LG, I am making no assumptions whatsoever about the manner of replication. Your account would make no difference.
I agree entirely that the space of possible intelligences is much wider than the intelligence that we are accustomed to. This simply allows for more variation, and so more likelihood for evolving to extinction, or again for evolving into conscious fitness maximizers.
"Evolved to Extinction" because female mice become rarer and rarer? "Female mice become rarer and rarer" is another way of saying at least 50% of all the genes in all the female individuals will make it to the next generation. Which is pretty damn good odds. Consider all the mutations in all those male individuals that will never get a chance to make it to the next generation, because the male individuals will never even get a chance to get close to a female, much less mate with one.
But the male chromosome isn't competing against the female chromosome. The mutant male chromosome is competing against the unmutant male chromosome. The mutant male chromosome is fitter, rises to fixation at its allele location, and in one more generation the species as a whole goes extinct.
Even if we decide that "evolving to extinction" should not refer to any case where environmental conditions change - and I think it's fair when the environmental conditions change as a direct result of the evolution itself, rather than exogenously - it is still possible for a species to evolve to extinction directly.
Unknown, it's physically possible to have a population with internal variation and replication while ensuring that it doesn't fall into certain destructive patterns with specialized controls. This is probably not a particularly good example since afaik multicellular organisms don't normally have much variation in their genetic code, but one of the methods for "outlawing evolution" (as mentioned by Eliezer) in multicellular organisms is Macrophages attacking tumor cells. It's not perfect, of course; individuals still die from cancer. But that mechanism is something produced by natural selection; an intelligent designer could do much better. You haven't yet convincingly argued that variation and replication necessarily lead to destructive runaway effects; you might need controls (which natrual selection may never come up with) to prevent it, but that doesn't make it impossible.
"Leaving aside crazy countries that use artificial gender selection technologies."
What happens when we can select the baby's gender (and a whole bunch of other stuff) before birth, using genetic engineering? By transhumanist standards, this is a very conservative prediction; I don't think it's "crazy" to develop or use such technology.
"Humanity is facing multiple species-level extinction threats right now, and I gotta tell ya, there ain't a lot of people steppin' forward."
In comparison to other groups with similar membership levels, we're doing very well. The National Space Society has more than twenty thousand paying members. They have an annual budget of ~$1000K. How much has SIAI raised so far this year, off a donor base of a few hundred people? $500K? $600K?
"We don't have a way to precisely copy macroscopic objects yet, but that might change with further technological development. Do you think that Molecular Nanotechnology is fundamentally impossible? If so, why?"
This is a giant cheesecake fallacy. We could, in the future, create a society of beings which are identical down to the last bit. I, and I suspect most other people, would find such a society highly undesirable.
But the male chromosome isn't competing against the female chromosome. The mutant male chromosome is competing against the unmutant male chromosome. The mutant male chromosome is fitter, rises to fixation at its allele location, and in one more generation the species as a whole goes extinct.
I would still be loath to call it "evolved to death". Where is the "evolution"? You are describing an event that would wipe out a species in an instant (considering it on the time scales that evolution acts on). Species die out instantaneously (on an evolutionary time scales) for many reasons.
How else can I respond to an event that takes "one more generation" to kill the whole species? Nothing "evolved"; the species died because the evolved machinery of genes didn't preclude such a mutation from killing a species in a handful of generations. Too bad, so sad. If there was an "evolution fairy", she would have designed a better machinery of genes. But if it is essential for the preferred use of the phrase "evolved to death" to describe events that take place instantaneously on an evolutionary time scale, I have to describe that phrase as misleading.
it is still possible for a species to evolve to extinction directly.
Favor me with another example. I found the other examples lacking.
From the fog of my misunderstanding, I am surprised you would use the phrase "evolved to death" without it immediately being followed by qualifications and clarifications. I look forward to you removing this fog away from my person.
Eliezer,
"Humanity is facing multiple species-level extinction threats right now," Hasn't that been true for the entire history of humanity (asteroids, supervolcanoes, someone inventing agriculture and setting us on the path to creating destructive AI)?
Tom,
That's a very conservative prediction indeed: parents already use sperm sorting to select the sex of their children before conception.
This is a giant cheesecake fallacy. We could, in the future, create a society of beings which are identical down to the last bit. I, and I suspect most other people, would find such a society highly undesirable.
I reject the claim of committing a GCF. My statement was a reply to Unknown's claim that "no two physical things will ever be exactly alike.", which appeared in an argument that was specifically not restricted to humans, biological beings or intelligent entities. For this reply I was thinking more along the lines of "MNT assembler" than ">= human-level intelligence". If you build a technological infrastructure using small self-replicating machines, you probably don't want them to acquire random mutations without shutting down.
The countries I meant to designate as "crazy" are the ones using selection technologies to produce all boys. 1.36 boys per girl in some regions.
"if mice were to evolve the ability to fly and migrate for the winter, they would probably form a single reproductive population, and would evolve to extinction as the segregation-distorter evolved to fixation."
Doesn't this count as a case of group level selection? If you are correct in this supposition the populations where this allele becomes established are quickly wiped out. Presumably, since mice are ubiquitous, such populations are generally re-seeded by mice from other populations after an evolutionarily short time. Even if mis-segregation allele occurs fairly frequently and benefits from a fitness advantage any randomly chosen mouse in the world will not usually possess it. I'm sure that there are ways to define "group selection" so that this doesn't count, but I'd like to know what definition is being proposed before hand. Certainly this is a case of an equilibrium where a selectively favored allele generally doesn't exist because the populations in which it does exist die out. How does this differ from the case of viruses?
Simply put, it doesn't.
Some viruses are so extraordinarily lethal that they've never managed to spread throughout the human population - see ebola, which I believe has approximately a 90% lethality, and even the survivors are severely brain-damaged - because they harm their victims so rapidly that they don't have a chance to spread.
Our modern transportation systems have effectively eliminated most of the barriers between human populations. All of our eggs are in one basket. If a highly lethal virus that will spread throughout an entire population and kill it arises, that basket will be dropped.
'I would still be loath to call it "evolved to death". Where is the "evolution"?'
It happens because of a change in gene frequencies. That's what evolution is defined as -- changes in gene frequency. The mutant allele spreads and takes over. The population dies, but that doesn't keep the mutant allele from taking over among survivors while there are survivors.
Someone said the mutant allele isn't competing with X chromosomes. It's competing with both X and Y alleles. The mice start out with on average 3 X alleles per Y allele. At the end it's 1 X allele per mutant Y allele.
'Too bad, so sad. If there was an "evolution fairy", she would have designed a better machinery of genes.'
There may be a better machinery sometimes. The examples we have of segregation distorters may be examples where the better mechanism has broken down.
In the absence of better mechanisms, this sort of thing can happen. It's possible to get mutations which are selected and which spread even though in the long run they are bad for the population as a whole.
Analogously it's possible to make a lot of money in a free enterprise system while benefitting nobody but yourself, barring mechanisms to prevent that -- the "invisible hand" doesn't necessarily work to do what Adam Smith said it can do.
"Doesn't this count as a case of group level selection?"
Yes, when it works. Divide the population into smaller groups with strictly limited breeding between groups, and that's one way that segregation distorters can be limited.
Better mechanisms might arise too, but until they do this is what you've got. There might be some other advantages to a population divided into small groups with limited interbreeding, too.
And once you have a population divided that way, it leaves the possibility for other group selection. However, rats are mammals, and mammals are a small minority group that are unimportant in the bigger scheme of things. How often are species divided up like that?
'Our modern transportation systems have effectively eliminated most of the barriers between human populations. All of our eggs are in one basket. If a highly lethal virus that will spread throughout an entire population and kill it arises, that basket will be dropped.'
This is a strong argument to change that situation. We have a communications system that lets us transmit data widely without needing personal contact. We could do some sorts of trade without a whole lot of risk, and minimise both the risk and the amount of trade for the rest.
It would be hard to do that effectively without a consensus among most of the world population, since people are so good at sneaking. So ideally over the next generation or so we'll develop a convincing case for dividing the world up into small breeding groups with strictly limited contact among groups, assuming the case actually is convincing.
At the moment the idea seems almost fantastical, as if we'd never put up with something like that. So we have a long way to go.
I bet the most widely used "selection technology" for producing boys rather than girls is infanticide, probably followed at a distance by (ultrasound scans followed by) abortion. I'm disinclined to call those "selection technologies" myself, but I suppose terminology varies.
The issue is replication with variation and the necessary historical consequences of this.
Evolution requires more than replication with variation. It needs differential replication with variation.
There is therefore no way to avoid the consequences of evolution: they are not biological consequences, but consequences of the laws of physics and logic. There is no way around them.
I can think of a couple of potential ways to avoid the consequences of evolution, by attacking the "differential" part.
The Singleton.
Some other method for achieving absolute security and property rights. For example a completely impenetrable shield. Or having automatic fail-proof self-destruct mechanisms built into everything to make it pointless for anyone to try to appropriate other people's property.
virii
Please don't use this non-word. It isn't correct, either in English or Latin, and just marks users of it as stupid and uneducated (which you are not).
Imagine a "Frodo gene" that sacrifices its vehicle to save its entire species from an extinction event. What happens to the allele frequency as a result? It goes down. Kthxbye.
But if 1 sacrifice ensures the survival of the offspring, then the genes that made the creature commit suicide would have still "won".
Suicide is not even required, just extreme dedication - putting your life on the line. Think motherhood - a mother bear will willingly fight to the death with a bear far bigger than her in order to protect her cubs. Sometimes this leads to her getting killed, but more than often it leads to the survival of the cubs.
Another good example of this type of behaviour is the pecking order; certain individuals submit to the "authority" of others, giving up food, mating rights and fill their body with stress hormones that weaken the immune system.
Shouldn't have pecking orders evolved out of the species and leave us with macho males that never submit?
You seem to imply there's a contradiction between "the good of the species" and "the selfishness of the gene", but it seems you're not paying attention to the subtle ways in which selection works.
But if 1 sacrifice ensures the survival of the offspring, then the genes that made the creature commit suicide would have still "won".
How so? They'd disappear. Unless the individual had already reproduced, as in your mother bear example - that's the difference. If Frodo did it to save his children, it might stay steady.
I think the important thing is that the critter with the frodo gene is not disproportionately saving other critters with the same gene relative to ones without it.
Virii: Please don't use this non-word.
Oh, fine. Hmph.
Can you recommend more in-depth references?
Depending on what you're looking for, either pick up a random textbook on evolutionary biology with math, or George Williams's Adaptation and Natural Selection.
For example, there is a segregation-distorter on the male sex chromosome of some mice which causes only male children to be born, all carrying the segregation-distorter.
Which locus is this?
And how does this relate to the jibber-jabber that is Darwin's _On_the_Origin_ofSpecies? I can't find the word "gene" in the index.
The concept of evolution has, uhh, evolved after Darwin first came up with it. You can't find the content of the latest peer-reviewed articles in evolutionary biology journals in Origin of Species, either.
Eliezer,
Humanity is facing multiple species-level extinction threats right now, and I gotta tell ya, there ain't a lot of people steppin' forward.
While I am hesitant to be part of the problem by sounding like an apologist or offering rationalizations, I have personal reasons for hope. Fittingly, one of the reasons for hope is that the process of evolution/emergence seems to introduce and sustain latent heterogeneity -- in the gene pool, in the idea pool, etc. This heterogeneity acts as a hedge, making the evolving population of agents more robust as a system. Burton Vorhees defines this as "virtual stability" and his recent work models it generally. So, even while there is reason for hope, this should not stop us from taking the threats you allude to seriously and actively seeking to eliminate those threats. Such vigilance is the diversity/hedge in the memeosphere against extinction due to complacency.
What is missing from this discussion is the complementary dynamic to natural selection, often called emergence, which is responsible for creating new, higher levels of system organization. As Stuart Kauffman argues, one of the ways emergence happens is through a generalized process of autocatalysis. But it also comes about through other means, namely cooperation of agents at lower levels.
When a new level emerges, given that it has yielded a population of higher-level agents (as opposed to just one or a few), and given that the population has the feature of agent replication with differential fitness, then natural selection (NS) occurs at the higher level. But NS does not stop at the lower level(s), it continues, as you point out for example in the case of populations of cells in the body of multicellular organisms (which can lead to cancer), populations of alleles (which include transposons), and so on.
NS occurs at all levels simultaneously as long as the preconditions of the dynamic are met. However, it is the case that stability of the higher level partially depends on the constraining NS at constituent lower levels. As you point out, multicellular organisms can only exist because they've evolved powerful internal mechanisms to outlaw evolution. If the cells start evolving, they rapidly evolve to extinction: the organism dies. Similarly, in the population of ideas/memes, the higher level of culture and ideology constrains which memes survive and thrive.
While nothing says that higher levels always emerge and that lower levels can't become unstable and devolve, but the tendency is for new levels to emerge and over time and constrain activity that would be destructive to the higher level agents. The process is inherently non-linear and volatile, and yes we may destroy ourselves before higher levels constrain us from doing so, but to me the trajectory looks promising. As Steven Pinker points out, violence has been in decline since recorded history, contrary to popular belief.
If we do survive the existential threats, what worries me more (as a lower-level agent) is the effect of higher level constraints on the human spirit and individual fulfillment. What's good for the organization isn't necessarily good for the individuals within the organization. We see examples of this everywhere, from small, loose communities to multinational corporations, governments, religions and ideologies. And as higher-level structures become more complex and "real", it will be harder for us as individuals to assert our rights to liberty and the pursuit of happiness.
For the record: I did some Googling, and I found out that the "t-complex" segregation distorter in mice causes various fitness penalties (such as impaired male fertility) in addition to manipulating the sex ratio. Males without the gene actually are more likely to successfully reproduce than males without the gene, so it never reaches an equilibrium of 100%.
Re: the battle between two alternative alleles at the same location is a zero-sum game
Yes, but there's also duplication and transposition to consider, so - in general - genetic patterns do not face simple zero-sum games - even with what seem-like their alleles.
Evolving to extinction nearly happened to Homo sapiens. This huge peacock's tail of a brain - which we grew just to do chimpanzee tribal politics better - really, that was the entire reason - causes a massive metabolic drain and makes our reproduction horribly difficult, frequently killing the mother and leaving infants helpless for years and with a horrible early mortality rate.
If the bottleneck 50,000 years ago had killed all of us instead of leaving a couple of thousand, we'd be a perfect example for the next intelligent species to use.
We're just lucky that evolution came up with general intelligence as a solution to convincing that other chimp to give you the fruit, and that general intelligence turned out to be quite useful for some other things.
really, that was the entire reason
In this spirit, I frequently remind myself, when the difficulties of dealing with other apes are getting to be too much to tolerate, that it's a problem so hard that we evolved brains capable of maintaining a global technological civilization and figuring out quantum mechanics as a side effect of trying to solve it, and I should maybe cut myself some slack.
Hard to say if politics was the entire reason... We are also endurance hunters and trap layers and both of those require being able to predict what your intended prey will do many steps in advance...
Question is, which came first?
And really, evolution didn't come up with a general intelligence to solve ape politics. Pay attention when you're thinking about things. How often do you reflexively think of inanimate objects as "wanting" or "happy"? You're probably modeling plants and animals and machines and complex physics as though it were another ape. Ape behaviour is so complex that other, complex systems can fit right into that rules processing engine, but that engine leaves its fingerprints all over the results...
Which is the entire reason this website exists. If we truly had a general-purpose intelligence most of the glitches in our thinking that we have to learn to be careful of wouldn't be there to start with.
You can't even praise evolution for the solicitous concern it shows for genes; the battle between two alternative alleles at the same location is a zero-sum game for frequency.
It seems like an artificially negative way of looking at it. Alleles are in competition with every other species. It seems rather biased to only pick out the other alleles at the same loci to compare with, and then to compare frequencies rather than absoulte allele numbers (which is not a zero-sum game).
Basically, this post is a rant against evoultion. Yet, evolution made you and me, and all living things on the planet. I think you should give credit where credit is due.
Suppose that in some sexually reproducing species, a perfect DNA-copying mechanism is invented. [...] The perfect DNA-copying mechanism rises in frequency to fixation. Ten thousand years later there's an ice age and the species goes out of business. It evolved to extinction.
This bit is surely not technically very likely. We have a range of species with different error-correction rates. It is pretty clear that the rate is largely under adaptive control - and that better and worse error correction mechanisms arise all the time.
High quality error correction has drawbacks - partly because mutation rate control requires energy and resources - and so it is not usually selectively favoured beyond a certain point.
Shit, man. I got no skin in the game as far as being a "real scientist" goes, or even studying evolution or biology and gene editing in a traditional sense, unless you count the unquantifiable amount of nights I suddenly look up at the clock and go
"god damn it. it's 430am, and I have work tomorrow. at 7am. WHY the entire and actual fuck did I just spend the last 6 hours reading about CRISPR, evolution, and advanced nuclear theory?"
that said, I got here because it's taken me this long, from when I first started looking this shit up when I was about 22 or 23, until now at age 35, to think:
"hmmm. I wonder if there's a confirmed case of a species evolving to be so efficient in it's environment, that the need to be good at sometime, or strong, or fast, or whatever, that it allows for mass complacency, which, over time, resulted in total extinction."
this was the first Google result from putting in "has a species ever evolved to extinction scientifically confirmed". and I'm not disappointed. it's so fucking refreshing to read something like this instead of what I see on my Twitter feed. god damn I miss smart people.
thanks for keeping this up on the web and paying for it's hosting for, judging by the comments, over 15 years. we need more of this.
cheers from California.
It's speculated that real-world group selection may have played a role in keeping the frequency of this gene as low as it seems to be
Super Important - it seems like we need silos, subgroups for evolution to work best. Since multi-level selection seems to be the way of the world, we need to give it all the levels it need to act on to make sure it acts most effectively. This could be a useful argument in a company for keeping teams relatively small, but then having the team leaders communicate and coordinate.
It is a very common misconception that an evolution works for the good of its species. Can you remember hearing someone talk about two rabbits breeding eight rabbits and thereby "contributing to the survival of their species"? A modern evolutionary biologist would never say such a thing; they'd sooner breed with a rabbit.
It's yet another case where you've got to simultaneously consider multiple abstract concepts and keep them distinct. Evolution doesn't operate on particular individuals; individuals keep whatever genes they're born with. Evolution operates on a reproducing population, a species, over time. There's a natural tendency to think that if an Evolution Fairy is operating on the species, she must be optimizing for the species. But what really changes are the gene frequencies, and frequencies don't increase or decrease according to how much the gene helps the species as a whole. As we shall later see, it's quite possible for a species to evolve to extinction.
Why are boys and girls born in roughly equal numbers? (Leaving aside crazy countries that use artificial gender selection technologies.) To see why this is surprising, consider that 1 male can impregnate 2, 10, or 100 females; it wouldn't seem that you need the same number of males as females to ensure the survival of the species. This is even more surprising in the vast majority of animal species where the male contributes very little to raising the children—humans are extraordinary, even among primates, for their level of paternal investment. Balanced gender ratios are found even in species where the male impregnates the female and vanishes into the mist.
Consider two groups on different sides of a mountain; in group A, each mother gives birth to 2 males and 2 females; in group B, each mother gives birth to 3 females and 1 male. Group A and group B will have the same number of children, but group B will have 50% more grandchildren and 125% more great-grandchildren. You might think this would be a significant evolutionary advantage.
But consider: The rarer males become, the more reproductively valuable they become—not to the group, but to the individual parent. Every child has one male and one female parent. Then in every generation, the total genetic contribution from all males equals the total genetic contribution from all females. The fewer males, the greater the individual genetic contribution per male. If all the females around you are doing what's good for the group, what's good for the species, and birthing 1 male per 10 females, you can make a genetic killing by birthing all males, each of whom will have (on average) ten times as many grandchildren as their female cousins.
So while group selection ought to favor more girls, individual selection favors equal investment in male and female offspring. Looking at the statistics of a maternity ward, you can see at a glance that the quantitative balance between group selection forces and individual selection forces is overwhelmingly tilted in favor of individual selection in Homo sapiens.
(Technically, this isn't quite a glance. Individual selection favors equal parental investments in male and female offspring. If males cost half as much to birth and/or raise, twice as many males as females will be born at the evolutionarily stable equilibrium. If the same number of males and females were born in the population at large, but males were twice as cheap to birth, then you could again make a genetic killing by birthing more males. So the maternity ward should reflect the balance of parental opportunity costs, in a hunter-gatherer society, between raising boys and raising girls; and you'd have to assess that somehow. But ya know, it doesn't seem all that much more reproductive-opportunity-costly for a hunter-gatherer family to raise a girl, so it's kinda suspicious that around the same number of boys are born as girls.)
Natural selection isn't about groups, or species, or even individuals. In a sexual species, an individual organism doesn't evolve; it keeps whatever genes it's born with. An individual is a once-off collection of genes that will never reappear; how can you select on that? When you consider that nearly all of your ancestors are dead, it's clear that "survival of the fittest" is a tremendous misnomer. "Replication of the fitter" would be more accurate, although technically, fitness is defined only in terms of replication.
Natural selection is really about gene frequencies. To get a complex adaptation, a machine with multiple dependent parts, each new gene as it evolves depends on the other genes being reliably present in its genetic environment. They must have high frequencies. The more complex the machine, the higher the frequencies must be. The signature of natural selection occurring is a gene rising from 0.00001% of the gene pool to 99% of the gene pool. This is the information, in an information-theoretic sense; and this is what must happen for large complex adaptations to evolve.
The real struggle in natural selection is not the competition of organisms for resources; this is an ephemeral thing when all the participants will vanish in another generation. The real struggle is the competition of alleles for frequency in the gene pool. This is the lasting consequence that creates lasting information. The two rams bellowing and locking horns are only passing shadows.
It's perfectly possible for an allele to spread to fixation by outcompeting an alternative allele which was "better for the species". If the Flying Spaghetti Monster magically created a species whose gender mix was perfectly optimized to ensure the survival of the species—the optimal gender mix to bounce back reliably from near-extinction events, adapt to new niches, etcetera—then the evolution would rapidly degrade this species optimum back into the individual-selection optimum of equal parental investment in males and females.
Imagine a "Frodo gene" that sacrifices its vehicle to save its entire species from an extinction event. What happens to the allele frequency as a result? It goes down. Kthxbye.
If species-level extinction threats occur regularly (call this a "Buffy environment") then the Frodo gene will systematically decrease in frequency and vanish, and soon thereafter, so will the species. A hypothetical example? Maybe. If the human species was going to stay biological for another century, it would be a good idea to start cloning Gandhi.
In viruses, there's the tension between individual viruses replicating as fast as possible, versus the benefit of leaving the host alive long enough to transmit the illness. This is a good real-world example of group selection, and if the virus evolves to a point on the fitness landscape where the group selection pressures fail to overcome individual pressures, the virus could vanish shortly thereafter. I don't know if a disease has ever been caught in the act of evolving to extinction, but it's probably happened any number of times.
Segregation-distorters subvert the mechanisms that usually guarantee fairness of sexual reproduction. For example, there is a segregation-distorter on the male sex chromosome of some mice which causes only male children to be born, all carrying the segregation-distorter. Then these males impregnate females, who give birth to only male children, and so on. You might cry "This is cheating!" but that's a human perspective; the reproductive fitness of this allele is extremely high, since it produces twice as many copies of itself in the succeeding generation as its nonmutant alternative. Even as females become rarer and rarer, males carrying this gene are no less likely to mate than any other male, and so the segregation-distorter remains twice as fit as its alternative allele. It's speculated that real-world group selection may have played a role in keeping the frequency of this gene as low as it seems to be. In which case, if mice were to evolve the ability to fly and migrate for the winter, they would probably form a single reproductive population, and would evolve to extinction as the segregation-distorter evolved to fixation.
Around 50% of the total genome of maize consists of transposons, DNA elements whose primary function is to copy themselves into other locations of DNA. A class of transposons called "P elements" seem to have first appeared in Drosophila only in the middle of the 20th century, and spread to every population of the species within 50 years. The "Alu sequence" in humans, a 300-base transposon, is repeated between 300,000 and a million times in the human genome. This may not extinguish a species, but it doesn't help it; transposons cause more mutations which are as always mostly harmful, decrease the effective copying fidelity of DNA. Yet such cheaters are extremely fit.
Suppose that in some sexually reproducing species, a perfect DNA-copying mechanism is invented. Since most mutations are detrimental, this gene complex is an advantage to its holders. Now you might wonder about beneficial mutations—they do happen occasionally, so wouldn't the unmutable be at a disadvantage? But in a sexual species, a beneficial mutation that began in a mutable can spread to the descendants of unmutables as well. The mutables suffer from degenerate mutations in each generation; and the unmutables can sexually acquire, and thereby benefit from, any beneficial mutations that occur in the mutables. Thus the mutables have a pure disadvantage. The perfect DNA-copying mechanism rises in frequency to fixation. Ten thousand years later there's an ice age and the species goes out of business. It evolved to extinction.
The "bystander effect" is that, when someone is in trouble, solitary individuals are more likely to intervene than groups. A college student apparently having an epileptic seizure was helped 85% of the time by a single bystander, and 31% of the time by five bystanders. I speculate that even if the kinship relation in a hunter-gatherer tribe was strong enough to create a selection pressure for helping individuals not directly related, when several potential helpers were present, a genetic arms race might occur to be the last one to step forward. Everyone delays, hoping that someone else will do it. Humanity is facing multiple species-level extinction threats right now, and I gotta tell ya, there ain't a lot of people steppin' forward. If we lose this fight because virtually no one showed up on the battlefield, then—like a probably-large number of species which we don't see around today—we will have evolved to extinction.
Cancerous cells do pretty well in the body, prospering and amassing more resources, far outcompeting their more obedient counterparts. For a while.
Multicellular organisms can only exist because they've evolved powerful internal mechanisms to outlaw evolution. If the cells start evolving, they rapidly evolve to extinction: the organism dies.
So praise not evolution for the solicitous concern it shows for the individual; nearly all of your ancestors are dead. Praise not evolution for the solicitous concern it shows for a species; no one has ever found a complex adaptation which can only be interpreted as operating to preserve a species, and the mathematics would seem to indicate that this is virtually impossible. Indeed, it's perfectly possible for a species to evolve to extinction. Humanity may be finishing up the process right now. You can't even praise evolution for the solicitous concern it shows for genes; the battle between two alternative alleles at the same location is a zero-sum game for frequency.
Fitness is not always your friend.