The Origin of Life

Intelligent Design

The term "intelligent design" refers to design that is performed using intelligence. Naturally many huamn artifacts are the product of intelligent design by humans. There is - or there should be - a science of intelligent design that looks at how such artifacts are created. The term "intelligent design" is also often confusingly used a shorhand for the hypothesis that some aspect of life as we know it was designed by intelligent agents.

In the latter sense, the term is often linked to the views of those religious fundamentalists - who often believe that the world and man were created by some sort of supreme being. These folks have made a bad name for themselves in America, by insiting that their unlikely views are taught to children in schools.

As many atheists argue, intelligent design theories seem improbable - and most supported of intelligent design are motivated by outdated and incorrect religious dogma.

However, this doesn't mean that intelligent design has been falsified, or that intelligent design hypotheses are unscientific.

The remaining viable scientific hypotheses of intelligent design should be judged on their merits.

Viable Intelligent Design Hypotheses

What are the remaining viable scientific hypotheses of intelligent design? Here is my list of the most sensible intelligent design hypotheses that may help explain the origin of life or man:

  • Simulism

    Simulism is the idea that reality as we perceive it is an illusion - and the world as we know it is a simulation.

    While speculating about the world that could be containing this one is not easy, looking at our own universe, many simulations mimic the real world, and are performed for entertainment purposes, or sometimes educational ones. Some of the most common types of simulated entertainment involve games, movies and pornography.

    Another important class of simulations are historical ones. Intelligent agents must model the past in order to predict the future. This hypothesis suggests that important historical events are more likely to be the subject of simulations. In particular attention might focus on major evolutionary transitions.

    Simulating major evolutionary transitions might be of interest to future intelligent agents because it might throw light on what aliens they might encounter in the future - and this could be an event of critical importance to them.

    Another scenario involves simulating people in order to influence them. Imagine a prototype intelligent machine, seeking donations from humans to help construct it. The machine could state that if it was successful, it would simulate large numbers of prospective donors and then put them all in a hell-like state if they do not donate. The machine could make its threat credible by exhibiting its source code. Donors might then reason that they are likely to be simulations, and thus be motivated to donate. Simulating agents can influence their decisions by creating uncertainty in them regarding whether they are being simulated.

    Simulism has recently found some celebrity supporters. Nick Bostrom assigned a probability of 20% to simulism being true - and has a domain associated with the hypothesis. Jaan Tallinn has also come out in favor of simulism - e.g. see: "Why Now? A Quest in Metaphysics".

    The simulation could be a simulation of the visible universe known by science - or it could be a simulation of one person's personal universe.

  • Optimisationverse

    The idea that the world is an optimisation algorithm is rather like Simulism - in that it postulates that the world exists inside a computer.

    However, the purpose of an optimisationverse is not entertainment - rather it is to solve some optimisation problem using a genetic algorithm.

    The genetic algorithm is a sophisticated one, that evolves its own recombination operators, discoveres engineering design - and so on.

    If this hypothesis is correct, it is not currently known what the problem to be solved is - or even whether it has been presented to anyone yet.

  • Panspermia

    Panspermia involves the idea that life on earth actually evolved elsewhere in the universe before making its way here. If that scenario is accurate, elements of life may have been engineered by intelligent designers on other worlds.

    The element of modern organisms most likely to have been designed in this way appears to have been the genetic substrate.

    As an example of a specific viable intelligent design hypothesis, consider the idea that the earth was seeded with RNA-based organisms about four billion years ago - and that these organisms originally came from a nearby star.

    What happened to the intelligent ancestors of these organisms? There are a number of possibilites - including the one discussed by Robin Hanson in his paper: Burning the Cosmic Commons:. Evolutionary Strategies for Interstellar Colonization.

    Richard Dawkins has discussed this idea recently Gods and earthlings. He writes: "Unrealistic as the space alien hypothesis is, it constitutes intelligent design's best shot." The latter point is debatable.

  • The adapted universe

    Lastly there is the possibility that the laws of physics are an adaptation. The laws do seem to exhibit "adaptive fit" - though this can be partly explained on anthropic grounds.

    This idea is associated with Lee Smolin: The Life of the Cosmos. Lee presents an unworkable version of the idea based on black holes.

    There is also the work of by James Gardner: Biocosm and Intelligent-Universe.

    In these books Gardner proposes that our universe has been deliberately engineered to promote life and intelligence - and requires these attributes in order to mediate the reproduction of the cosmos.

    This scenario places life in the universe as the product of previous intelligent agents, in our universe's ancestor.


The evidence in favor of simulation includes the following:

  • Universe seems to have been born - an endless universe would offer less scope for a creator.
  • The universe exhibits microscopic reversibility. This property would be useful, if someone was paying for its power supply and heat dissipation requirements. Reversible physics is not necessary for life but reversibility helps with rewind functionality and backtracking to explore alternative paths.

  • The universe appears to be more life-friendly than is required by anthropic arguments. It looks a lot as though living systems will expand to fill the universe - using most of the available matter in one way or another. This is compatible with living systems being a focus of the intelligent designers.

  • The universe's net energy appears to be zero. This makes sense if one assumes that someone had to pay for the resources needed to construct the universe.

  • The universe's initial conditions appear to have been very simple. Also, the laws of physics appear to be simple, local and uniform. Again, this all makes sense if one assumes that someone had to pay for setting up the initial conditions.

  • These are exciting times - ancestor simulations could easily focus on major evolutionary transitions.

Some of these observations are suggestive - but they are far-from conclusive. There is not much evidence against these hypotheses either - absence of evidence is only weak evidence of absence.

Probability assessments

Richard Dawkins - never seems to give an concrete probability estimate - but he has said that he assigns a "very low" probability to the hypothesis that god exists.

Tyler Cowen has argued that many atheists put too low a probability on the possibility that intelligent design was involved in the origin of the universe life or man. He wrote:

I've asked atheists what's the chance you're wrong and they'll say something like a trillion to one, and that to me is absurd, that even if you think all of the strongest arguments for atheism are correct, your estimate that atheism is in fact the correct point of view shouldn't be that high, maybe you know 90-10 or 95 to 5, at most.
Tyler Cowen's comment seems to me to be a correct assessment, reflecting the common problem of overconfidence. This would make the estimates of Richard Dawkins and Michael Shermer overconfident. This would not be surprising since these folk are famous for being atheists - so it would make sense for them to have lower estimates of p(god) than are scientifically defensible. Elon Musk and Hans Moravec also seem wildly overconfident - but in the opposite direction.

When giving probability estimates, some care needs to be taken regarding what probability is being estimated. Judging by today's simulations many simulated entities never figure out that they are living in simulations. From their point of view, simulism remains an unproven hypothesis. David Chalmers attempts to avoid this by saying that he gets to make a bet and god decides it - but this is assumung the existence of god - and that's the topic of the bet. This does not seem very scientific. I think we have to confine probability assessments to cases where evidence of the simulation accrues within the simulation - and bets can actually be decided.

My own estimate of the chance of one these hypotheses about intelligent design proving to be correct is around 10%. I reason that it is very challenging to decide what fraction of humans are simulated, since this requires a lot of speculation about the chances of superintelligent machines forming and being interested in simulating the human era. Because of these problems I figure that estimates much smaller than 10% would be overconfident.

Other celebrity endorsements


Intelligent Design has been subject of a lot of criticisms. Some of these criticisms apply to the theories on this page:

  • Q. The design hypothesis requires a creator god - isn't that more complex than evolution by natural selection.

    A. God could have evolved in the base universe. The complexity of this hypothesis seems broadly similar to us being in the basement universe.

  • Q. Why did god take such pains to hide himself?

    A. Simulism offers many possible answers to this. Fiction characters rarely seem to be aware that they are fictional. Scientists conceal their presence from their subjects, to avoid bias. If simulation is intended to influence the subject the simulation should be of similar quality to the original.

  • Q. The design hypothesis immediately raises the larger problem of: who designed the designer?

    A. The designer could have been designed, or they could have evolved via conventional evolutionary mechanisms including natural selection.

  • Q. Doesn't putting origin questions into an enclosing universe make them hard for science to answer?

    A. Yes, but that's where the actual answer might be. There's no rule that says that origin issues must be easily accessible to science.

Simpler simulations

Simulism is probably the most likely intelligent design scenario. Probably the strongest criticism of it involves doubting the claim that simulated humans are common - relative to original ones. We currently see that most simulations are simpler than the original systems they simulate. In that case, if we are simulations, these are likely to be taking place in a world which is larger and richer than our own. Larger worlds are probably on average slightly more complex than smaller ones - and thus slightly less common. Opposing this is the tendency for organisms to produce large numbers of simulations in their minds. For example, if you look at the chess positions considered by chess computers, only a tiny fraction of these ever exist on an acutal game board. The rest are simulated positions. The chance of simulism being true may depend on the balance between these opposing tendencies. The answer depends on the probability that universes that evolve as far as humans go on to evolve powerful superintelligences. Unfortunately, this is not an easy sum to calculate since we lack data relating to it.

Occam's razor

Eray Ozkural has produced some criticism of simulism - in: Why the Simulation Argument is Invalid. He says simulism involves creationism and solipsism. He claims a 2-99,000 chance of intelligent design creationism being true. Alas, his maths doesn't make much sense. He gives a p=2-1,000 chance of the universe and its seven billion human brains forming, but p=2-100,000 as a consevative estimate for one human genome - as a ballpark estimate for the probability of an intelligent designer such as god. Why god is so much more complex than all human civilization is not properly explained. If the universe consists of the execution of a 1,000-bit program, as Eray says, it might not require a rocket scientist to create it. Eray apparently neglects the possibility that god's brain evolved in a universe much like our own. If seven billion human brains can be the result of the execution of a 1,000 program, then so can advanced designers with the resources of a galaxy under their control. I directed Eray's attention to this issue, but he failed to give a coherent counter-argument.


Another criticism says that if there's a god, we might have no way to know it - in which case god's existence would remain opaque to science - and it wouldn't make any difference. That's a possibility - but not a very interesting one. In some cases where there's a god there's also a reasonable chance of its creations figuring that out.

Here's Jerry Coyne making that objection:

For one thing, they are untestable claims and therefore unscientific ones. How would we know that we're manipulated by aliens, or even part of a simulation? Further, it's unparsimonious. What reason do we have for thinking that we are a gigantic real or virtual experiment rather than inhabitants of a real Universe?
Eray Ozkural makes much the same objection. The issue is covered in the simulation argument FAQ.

Massimo Pigliucci writes:

I actually think the argument is interesting, philosophically speaking, but it has two problems: it is entirely empirically untestable - i.e., it's not science - and is based on what I think is a fundamental flaw.

Moral objections

Several of the critics of simulism make a moral argument: that the world is full of suffering and that simulating it would be immoral.

For example, here's Magnus Vinding on the topic in 2015:

A fully conscious simulation of our own past up until this point entails an unspeakable amount of horror and suffering: wars, famines, torture etc. – an ocean of suffering in which the horrors brought to the world by Hitler and Stalin are mere droplets. Any individual or civilization that would intentionally create all this suffering - holocaust upon holocaust of suffering - for the mere sake of curiosity would be evil to an extent that is unmatched in human history. Given our technological and moral progress up until this point in history, it seems very unlikely that we will ever create such an atrocity that dwarfs all other humanly created suffering by recreating it all, and far more on top of it.
I think that argument is probably mistaken. Intelligent agents build models of the world and test them against existing knowledge. Having an inaccurate world model could cause a failure to thrive or survive. Making sure that the world model is accurate by testing it against historical data seems likely to trump ethical concerns about running historical simulations - since having an inaccurate would model could damage to your entire future timeline - and that is quite important.

David Pearce argues that:

The existence of suffering is strong presumptive evidence that our descendants will never run sentience-supporting ancestor-simulations.


Nick Bostrom's "simulation argument" has been criticized for not being "multiverse aware". That seems like a reasonable criticism to me. However, some critics seem to think multiverse considerations largely invalidate the arument. In particular, Brian Eggleston writes:

But, it is clear that the probability P(W) is simply the prior probability that we place on the existence of a world other than our own. If this probability is taken to be very small, then the conclusion of the simulation argument doesn't follow, and we cannot conclude that it is probable that we are living in a computer simulation.
I identify the problem with Brian's argument as being in this passage:

However, obviously we cannot count individuals from simulations that we ourselves run, because these simulated individuals don't contribute to the possibility that we are in a simulated universe, since we know for sure that we are not them, since we created them. In fact, that only simulated individuals that can contribute the probability that we are living in a simulated universe are individuals that we haven't (and will not) create.
This idea is mistaken. Whether we create the sims or not is pretty irrelevant to the topic.


Robin Hanson argues against simulism here. He argues that future folk probably won't be very interested in running full-scale ancestor simulations. He gives a probability estimate at the end of the article - but it is a rather vague one. He says his chances of being in a simulation are less than 50%.

Does it matter?

One argument against funding more research into the possibility of intelligent design is that the results of such research would not matter. For example if we found out that we were inside a massive simulation, it would make no difference to our behaviour. This is indeed a possibility. However, if we are in a simulation, maybe it really would matter what we do, and will influence decisions in the future. The idea that evidence for intelligent design would not affect our actions is fairly speculative - we should not avoid such research on these grounds.


This page is devoted to a scientific approach to intelligent design of man, life, or the universe. However, most enthusiasts for intelligent design are not scientists, but people seeking to justify their religious beliefs, or pseudoscientists. Many of these deny the validity of Darwin's theory of evolution, which is a very well established scientific theory. The existence of all these confused individuals is unfortunate. Maybe some scientists will consider avoiding the same terminology that they use - and avoid the terms "creationism" and "intelligent design". The terms do seem to be quite appropriate, though. Perhaps scientists will use them after all.

Criticizing the wrong theories

Critics of intelligent design are common. However most of them concentrate their attention on the intelligent design pseudoscientists. These are easy targets; it seems like shooting fish in a barrel. From a scientific perspective, criticizing the weakest versions of a theory is a futile activity that rarely advances knowledge. Instead, critics should seek out the strongest version of a theory - creating it if necessary - a process often known as "steelmanning".

Tim Tyler | Contact |