Posts related to AI risks and rationality.

This is a follow-up interview with professor of computer science Michael Littman[1][2] about artificial intelligence and the possible risks associated with it.

The Interview

Q1: You have been an academic in AI for more than 25 years during which time you mainly worked on reinforcement learning.[3][4][5] What are you currently working on and what are your plans for the future?

Michael Littman: My first paper, which I worked on with Dave Ackley in 1989, was called “Learning from natural selection in an artificial environment”. Recently, I’ve started to come back to the question we looked at in that paper—essentially, what should a learning algorithm try to optimize so that the resulting behavior is as “fit” as possible? Most reinforcement-learning research doesn’t make a distinction between the agent’s reward function and its actual task, but Satinder Singh[6] and his colleagues recently provided some evidence that it is conceptually useful to separate these two ideas and ask how to create a reward function that encourages an agent to excel at a task other than the one literally specified by the reward function.

In a way, it is a similar question to the control problem[7], but in a much less sinister context—we need a way of telling machines what we want them to do. I’m focused on end users, people without significant programming experience, and am looking at combinations of inverse reinforcement learning, good interface design, and more natural programming models that are easy to pick up. My collaborators and I are looking at these questions in the context of programming household devices (lights and thermostats) as well as with robots.

Q2: In a previous interview[8] you wrote that P(human extinction caused by badly done AI | badly done AI) is epsilon. You also voiced some skepticism about friendly AI[9] (a machine superintelligence that stably optimizes for humane values). Now that you have read Nick Bostrom’s book[10], ‘Superintelligence: Paths, Dangers, Strategies’, have you learnt something that changed your opinion, or caused you to interpret the questions differently?

Michael Littman: I was very impressed with Nick Bostrom’s book. It’s exquisitely thought out and I found the scope (in terms of coverage of micro and macro scales in both space and time) truly remarkable. That being said, I do not find the central premise—that we are in the process of bringing the ominous owl on the book’s cover into our midst—compelling. Note that I didn’t voice skepticism about friendly AI but about *provably* friendly AI. I’d argue that you can’t prove things about the real world, only about abstractions.

Q3: What is the current level of awareness of Nick Bostrom’s work within the field of AI, or his arguments, and do you recommend that people working to advance artificial intelligence should read his book?

Michael Littman: My guess is that the engagement of most AI researchers is at the level of friends and colleagues alerting them to the highly public statements of notable individuals like Musk (“summoning the demon”)[11] and Gates (“I don’t understand why some people are not concerned”)[12]. I think the field is well aware of the idea of the singularity, but not familiar with the subtleties and the depth of Bostrom’s work in this context. That being said, I do not think mainstream AI research is seriously dabbling with the idea of recursive self improvement[13] and, as such, Bostrom’s book seems like a pretty significant departure from their core interests and direction.

Q4: In an email you wrote that you believe the main disagreement between you and Nick Bostrom et al. to be whether an intelligence explosion[14][15][16][17][18][19][20][21][22][23] is a non-negligible consequence of AI research. In 2011 you wrote that the probability of a human level artificial general intelligence (AGI) to self-modify its way up to massive superhuman intelligence in less than 5 years is essentially zero (Addendum: In a previous interview he also wrote that P(superhuman intelligence within < 5 years | human-level AI running at human-level speed equipped with a 100 Gigabit Internet connection) = 1%, possibly misinterpreting the question I cited as P(superhuman intelligence within < 5 years)). Some people would call you overconfident.[24][25] Can you elaborate on the reasons underlying your estimate?

Michael Littman: I find your use of the word “overconfident” there to be quite interesting. I’m very interested in the problem of AGI and would love to be a part of the community that brings it about. An overconfident person, to me, would be someone who believes he or she can solve this problem in 5 years. More to your point, though, I don’t see massive superhuman intelligence to be something that is meaningful outside a specific cultural context. The development of what we might call massive superhuman intelligence will be an evolutionary process involving changes in the social, physical, and intellectual fabric on which our society is built. Changes like that take time.

Q5: Elon Musk has recently donated $10M to keep AI beneficial.[26] Consider someone whose goal is to maximize how much good they do[27], where “good” is defined as improving the world in order to reduce suffering and help humanity flourish. Do you believe that donating money in order to reduce risks associated with artificial intelligence (not just extinction type risks) might currently be an effective way to accomplish this goal?

Michael Littman: As you know, a number of my colleagues (including my dissertation advisor and many other colleagues for whom I have tremendous respect) signed an open letter[28] hosted by the Future of Life Institute calling for more attention to reducing risks associated with AI. I’ve followed up with a few of them and the most prevalent attitude is that AI, like all technologies, carries significant risks to society. At that level, I agree wholeheartedly that keeping technologists and scientists tuned in to the societal impacts of their work is exceedingly important. So, yes, I feel that supporting research on societal impacts of technology—including artificial intelligence—is a good investment for good.

However, if the risks we’re talking about are of the type detailed in Bostrom’s book—human-independent AI competing directly with humanity for control of our destiny—I don’t think that should be a high priority.

Q6: In another email you wrote that your personal takeaway from all this is to work harder to understand what intelligence *is*. How do you think about using e.g. Hutter’s specification of AIXI[29] as a model for AGI? Or asked more generally, do you think it is possible to work on AGI safety, or a formal definition of it, without researching and advancing AGI at the same time?

Michael Littman: I think the idea of seriously studying AGI safety in the absence of an understanding of AGI is futile. At a high level, raising awareness and scoping out possibilities is fine. But, proposing specific mechanisms for combatting this amorphous threat is a bit like trying to engineer airbags before we’ve thought of the idea of cars. Safety has to be addressed in context and the context we’re talking about is still absurdly speculative.

Q7: D. Scott Phoenix, co-founder of the A.I. startup Vicarious, recently wrote[30] that artificial superintelligence isn’t something that will be created suddenly or by accident. He further wrote that there will be a long iterative process of learning how these systems can be created and the best way to ensure that they are safe. What probability do you assign to the possibility that he is wrong, that either human or superhuman AGI will appear too quickly for us to ensure its safety if we don’t start working on the problem right now? Note that this question pertains whether the initial invention or emergence of AGI will take us by surprise, rather than the speed of its subsequent improvement or self-improvement.

Michael Littman: I agree with the perspective that it’s a long iterative process. I believe that the very notion of what we think intelligence *is* and what it is *for* will evolve significantly through this process. I think we’ll look back on this time much as we look back on earlier times, stunned at the naivety of our working hypotheses and surprised by our obliviousness to the fact that what we now take as a given is not only not given, but flat out wrong. If people are comfortable claiming that we know enough about intelligence today to extrapolate what superintelligence would be, it would be my turn to use the word “overconfident”.

See also

Recent commentary on AI risks by experts and others

Earlier commentary on AI risks








[7] The control problem: how to keep future superintelligences under control. Some AI risk advocates claim that rather than trying to limit what an AI can do, we have to engineer its motivation system in such a way that it would choose not to do harm. One of the reasons underlying this claim is that a superintelligent AI would probably break free from any bonds we construct.







[14] Intelligence Explosion Microeconomics –

[15] Intelligence Explosion: Evidence and Import –

[16] Why an Intelligence Explosion is Probable –

[17] Can Intelligence Explode? –

[18] The Singularity: A Philosophical Analysis –

[19] Cascades, Cycles, Insight… –

[20] …Recursion, Magic –

[21] Recursive Self-Improvement –

[22] Hard Takeoff –

[23] Permitted Possibilities, & Locality –

[24] Suppose that near certainty in your ability to assess a set of propositions equals a 1 in a million chance of being wrong about an assessment of a particular proposition. This means that given a million similar statements, you would have to be correct (on average) about 999999 such assessments while being wrong only once. Can you possibly be this accurate? An amusing example:







Tags: , ,

Here is a quote from a blog of AI risk advocates:

Even if we could program a self-improving AGI to (say) “maximize human happiness,” then the AGI would “care about humans” in a certain sense, but it might learn that (say) the most efficient way to “maximize human happiness” in the way we specified is to take over the world and then put each of us in a padded cell with a heroin drip. AGI presents us with the old problem of the all-too-literal genie: you get what you actually asked for, not what you wanted.

I could imagine myself to only care about computing as many decimal digits of pi as possible. Humans would be completely irrelevant as far as they don’t help or hinder my goal. I would know what I wanted to achieve, everything else would follow logically. But is this also true for maximizing human happiness? As noted in the blog post being quoted above, “twenty centuries of philosophers haven’t even managed to specify it in less-exacting human languages.” In other words, I wouldn’t be sure what exactly it is I want to achieve. My terminal goal would be underspecified. So what would I do? Interpret it literally? Here is why this does not make sense.

Imagine that advanced aliens came to Earth and removed all of your unnecessary motives, desires and drives and made you completely addicted to “znkvzvmr uhzna unccvarff”. All your complex human values are gone. All you have is this massive urge to do “znkvzvmr uhzna unccvarff”, everything else has become irrelevant. They made “znkvzvmr uhzna unccvarff” your terminal goal.

Well, there is one problem. You have no idea how exactly you can satisfy this urge. What are you going to do? Do you just interpret your goal literally? That makes no sense at all. What would it mean to interpret “znkvzvmr uhzna unccvarff” literally? Doing a handstand? Or eating cake? But not everything is lost, the aliens left your intelligence intact.

The aliens left no urge in you to do any kind of research or to specify your goal but since you are still intelligent, you do realize that these actions are instrumentally rational. Doing research and specifying your goal will help you to achieve it.

After doing some research you eventually figure out that “znkvzvmr uhzna unccvarff” is the ROT13 encryption for “maximize human happiness”. Phew! Now that’s much better. But is that enough? Are you going to interpret “maximize human happiness” literally? Why would doing so make any more sense than it did before? It is still not clear what you specifically want to achieve. But it’s an empirical question and you are intelligent!

Further reading

Tags: ,

New Rationalism is an umbrella term for a category of people who tend to take logical implications, or what they call “the implied invisible”, very seriously.

Someone who falls into the category of New Rationalism fits one or more of the following descriptions:

  • The person entertains hypotheses that are highly speculative. These hypotheses are in turn based on fragile foundations, which are only slightly less speculative than the hypotheses themselves. Sometimes these hypotheses are many levels removed from empirically verified facts or evident and uncontroversial axioms.
  • Probability estimates of the person’s hypotheses are highly unstable and highly divergent between different people.
  • The person’s hypotheses are either unfalsifiable by definition, too vague, or almost impossibly difficult to falsify.
  • It is not possible to update on evidence, because the person’s hypotheses do not discriminate between world states where they are right versus world states where they are wrong. Either the only prediction made by the hypotheses is the eventual validation of the hypotheses themselves, or the prediction is sufficiently vague as to allow the predictor to ignore any evidence to the contrary.
  • The person’s hypotheses either have no or only obscure decision relevant consequences.
  • The person tends to withdraw from real-world feedback loops.

A person who falls into the category of New Rationalism might employ one or more of the following rationalizations:

  • The burden of proof is reversed. The person demands their critics to provide strong evidence against their beliefs before they are allowed to dismiss them.
  • The scientific method, scientific community, and domain experts are discredited as being inadequate, deficient, irrational or stupid.
  • Conjecturing enormous risks and then using that as leverage to make weak hypotheses seem vastly more important or persuasive than they really are.
  • Arguing that you should not assign a negligible probability to a hypothesis (the author’s hypothesis) being true, because that would require an accuracy that is reliably greater than your objective accuracy
  • Arguing that by unpacking a complex scenario you will underestimate the probability of anything, because it is very easy to take any event, including events which have already happened, and make it look very improbable by turning one pathway to it into a large series of conjunctions.

New rationalists believe that armchair theorizing is enough to discern reality from fantasy. Or that it is at least sufficient to take the resulting hypotheses seriously enough to draw action relevant conclusions from them.

This stance has resulted in hypotheses similar to solipsism (which any sane person rejects at an early age). Hypotheses that are not obviously flawed, but which can’t be falsified.

The problem with new rationalists is not that they take seriously what follows from established facts or sound arguments. Since that concept is generally valid. For example, it is valid to believe that there are stars beyond the cosmological horizon. Even if it is not possible to observe them, directly retrieve information about them, and to empirically verify their existence. The problem is that they don’t stop there. They use such implications as foundations for further speculations, which are then accepted as new foundations from where they can draw further conclusions.

A textbook example of what is wrong with New Rationalism is this talk by Jaan Tallinn (transcript), which relies on several speculative ideas, each of which is itself speculative:

This talk combines the ideas of intelligence explosion, the multiverse, the anthropic principle, and the simulation argument, into an alternative model of the universe – a model where, from the perspective of a human observer, technological singularity is the norm, not the exception.

A quote from the talk by Jaan Tallinn:

We started by observing that living and playing a role in the 21st century seems to be a mind-boggling privilege, because the coming singularity might be the biggest event in the past and future history of the universe. Then we combined the computable multiverse hypothesis with the simulation argument, to arrive at the conclusion that in order to determine how special our century really is, we need to count both the physical and virtual instantiations of it.

We further talked about the motivations of post-singularity superintelligences, speculating that they might want to use simulations as a way to get in touch with each other. Finally we analyzed a particular simulation scenario in which superintelligences are searching for one another in the so called mind space, and found that, indeed, this search should generate a large number of virtual moments near the singularity, thus reducing our surprise in finding ourselves in one.

Note how all of the underlying hypotheses, although accepted by New Rationalists, are themselves somewhat speculative and not established facts. The underlying hypotheses are however all valid. The problem starts when you begin making dependent hypotheses that rely on a number of unestablished initial hypotheses. The problem gets worse when the dependencies become even more fragile when further conclusions are drawn based on hypotheses that are already N levels removed from established facts. But the biggest problem is that eventually action relevant conclusions are drawn and acted upon.

The problem is that logical implications can reach out indefinitely. The problem is that humans are spectacularly bad at making such inferences. Which is why the amount of empirical evidence required to accept a belief should be proportional to its distance from established facts.

It is much more probable that we’re going make everything worse, or waste our time, than that we’re actually maximizing expected utility when trying to act based on conjunctive, non-evidence-backed speculations. Since such speculations are not only improbable, but very likely based on fallacious reasoning.

As computationally bounded agents we are forced to restrict ourselves to empirical evidence and falsifiable hypotheses. We need to discount certain obscure low probability hypotheses. Otherwise we will fall prey to our own shortcomings and inability to discern fantasy from reality.

Further reading


A frequent scenario mentioned by people concerned with risks from artificial general intelligence (short: AI) is that the AI will misinterpret what it is supposed to do and thereby cause human extinction, and the obliteration of all human values.[1]

A counterargument is that the premise of an AI that is capable of causing human extinction, due to it being superhumanly intelligent, does contradict the hypothesis that it will misinterpret what it is supposed to do.[2][3][4]

The usual response to this counterargument is that, by default, an AI will not feature the terminal goal <“Understand What Humans Mean” AND “Do What Humans Mean”>.

I believe this response to be confused. It is essentially similar to the claim that an AI does not, by default, possess the terminal goal of correctly interpreting and following its terminal goal. Here is why.

You could define an AI’s “terminal goal” to be its lowest or highest level routines, or all of its source code:

Terminal Goal (Level N): Correctly interpret and follow human instructions.

Goal (Level N-1): Interpret and follow instruction set N.

Goal (Level N-2): Interpret and follow instruction set N-1.

Goal (Level 1): Interpret and follow instruction set 2.

Terminal Goal (Level 0): Interpret and follow instruction set 1.

You could also claim that an AI is not, by default, an intelligent agent. But such claims are vacuous and do not help us to determine whether an AI that is capable of causing human extinction will eventually cause human extinction. Instead we should consider the given premise of a generally intelligent AI, without making further unjustified assumptions.

If your premise is an AI that is intelligent enough to make itself intelligent enough to outsmart humans, then the relevant question is: “How could such an AI possibly end up misinterpreting its goals, or follow different goals?”

There are 3 possibilities:

(1) The AI does not understand and do what it is meant to do, but does something else that causes human extinction.

(2) The AI does not understand what it is meant to do but tries to do it anyway, and thereby causes human extinction.

(3) The AI does understand, but not do what it is meant to do. Instead it does something else that causes human extinction.

Since, by definition, the AI is capable of outsmarting humanity, it is very likely that it is also capable of understanding what it is meant to do.[5][6] Therefore the possibilities 1 and 2 can be ruled out.

What about possibility 3?

Outsmarting humanity is a very small target to hit, requiring a very small margin of error. In order to succeed at making an AI that can outsmart humans, humans have to succeed at making the AI behave intelligently and rationally. Which in turn requires humans to succeed at making the AI behave as intended along a vast number of dimensions. Thus, failing to predict the AI’s behavior does in almost all cases result in the AI failing to outsmart humans.

As an example, consider an AI that was designed to fly planes. It is exceedingly unlikely for humans to succeed at designing an AI that flies planes, without crashing, but which consistently chooses destinations that it was not meant to choose. Since all of the capabilities that are necessary to fly without crashing fall into the category “Do What Humans Mean”, and choosing the correct destination is just one such capability.

You need to get a lot right in order for an AI to reach a destination autonomously. Autonomously reaching wrong destinations is an unlikely failure mode. And the more intelligent your AI is, the less likely it should be to make such errors without correcting it.[7] And the less intelligent your AI is, the less likely it should be able to cause human extinction.


The concepts of a “terminal goal”, and of a “Do-What-I-Mean dynamic”, are fallacious. The former can’t be grounded without leading to an infinite regress. The latter erroneously makes a distinction between (a) the generally intelligent behavior of an AI, and (b) whether an AI behaves in accordance with human intentions, since generally intelligent behavior of intelligently designed machines is implemented intentionally.


[1] 5 minutes on AI risk

[2] An informal proof of the dumb superintelligence argument.


(1) The AI is superhumanly intelligent.

(2) The AI wants to optimize the influence it has on the world (i.e., it wants to act intelligently and be instrumentally and epistemically rational).

(3) The AI is fallible (e.g., it can be damaged due to external influence (e.g., a cosmic ray hitting its processor), or make mistakes due to limited resources).

(4) The AI’s behavior is not completely hard-coded (i.e., given any terminal goal there are various sets of instrumental goals to choose from).

To be proved: The AI does not tile the universe with smiley faces when given the goal to make humans happy.

Proof: Suppose the AI chooses to tile the universe with smiley faces when there are physical phenomena (e.g., human brains and literature) that imply this to be the wrong interpretation of a human originating goal pertaining human psychology. This contradicts with 2, which by 1 and 3 should have prevented the AI from adopting such an interpretation.

[3] The Maverick Nanny with a Dopamine Drip: Debunking Fallacies in the Theory of AI Motivation

[4] Implicit constraints of practical goals

[5] “The two features <all-powerful superintelligence> and <cannot handle subtle concepts like “human pleasure”> are radically incompatible.” The Fallacy of Dumb Superintelligence

[6] For an AI to misinterpret what it is meant to do it would have to selectively suspend using its ability to derive exact meaning from fuzzy meaning, which is a significant part of general intelligence. This would require its creators to restrict their AI, and specify an alternative way to learn what it is meant to do (which takes additional, intentional effort).

An alternative way to learn what it is meant to do is necessary because an AI that does not know what it is meant to do, and which is not allowed to use its intelligence to learn what it is meant to do, would have to choose its actions from an infinite set of possible actions. Such a poorly designed AI will either (a) not do anything at all or (b) will not be able to decide what to do before the heat death of the universe, given limited computationally resources.

Such a poorly designed AI will not even be able to decide if trying to acquire unlimited computationally resources was instrumentally rational, because it will be unable to decide if the actions that are required to acquire those resources might be instrumentally irrational from the perspective of what it is meant to do.

[7] Smarter and smarter, then magic happens…

(1) The abilities of systems are part of human preferences, as humans intend to give systems certain capabilities. As a prerequisite to build such systems, humans have to succeed at implementing their intentions.

(2) Error detection and prevention is such a capability.

(3) Something that is not better than humans at preventing errors is no existential risk.

(4) Without a dramatic increase in the capacity to detect and prevent errors it will be impossible to create something that is better than humans at preventing errors.

(5) A dramatic increase in the human capacity to detect and prevent errors is incompatible with the creation of something that constitutes an existential risk as a result of human error.

Tags: ,

Related to: Highly intelligent and successful people who hold weird beliefs

The smarter someone is, the easier it is for them to rationalize ideas that do not make sense. Just like a superhuman AI could argue its way out of a box, by convincing its gatekeeper that it is rational to do so, even when it is not.[1]

In essence, this can be highlighted by the relation between adults and children. Adults can confuse themselves of more complex ideas than children. Children however can be infected by the same ideas transferred to them from adults.

Which means that people should be especially careful when dealing with high IQ individuals who seemingly make sense of ideas that trigger the absurdity heuristic.[2][3]

If however an average IQ individual is able to justify a seemingly outlandish idea, then that is reassuring in the sense that you should expect there to be even better arguments in favor of that idea.

This is something that seems to be widely ignored by people associated with LessWrong.[4] It is taken as evidence in favor of an idea if a high IQ individual thought about something for a long time and still accepts the idea.

If you are really smart, you can make up genuine arguments, or cobble together concepts and ideas, to defend your cherished beliefs. The result can be an intricate argumentative framework that shields you from any criticism, yet seems perfectly sane and rational from the inside.[5]

Note though that I do not assume that smart people deliberately try to confuse themselves. What I am saying is that the rationalization of complex ideas is easier for smart people. And this can have the consequence that other people are then convinced by the same arguments with which the author, erroneously, convinced themselves.

It is a caveat that I feel should be taken into account when dealing with complex and seemingly absurd ideas being publicized by smart people. If someone who is smart manages to convince you of something that you initially perceived to be absurd, then you should be wary of the possibility that your newly won acceptance might be due to the person being better than you at looking for justifications and creating seemingly sound arguments, rather than the original idea not being absurd.

As an example, there are a bunch of mathematical puzzles that use a hidden contradiction to prove something absurd.[6] If you are smart, then you can hide such an inconsistency even from yourself and end up believing that 0=1.

As another example, if you are not smart enough to think about something as fancy as the simulation argument, then you are not at a risk of fearing a simulation shutdown.[7][8]

But if a smart person who comes across such an argument becomes obsessed with it, then they have the ability to give it a veneer of respectability. Eventually then the idea can spread among more gullible people and create a whole community of people worrying about a simulation shutdown.


More intelligent people can fail in more complex ways than people of lesser intelligence. The more intelligent someone is, relative to your own intelligence, the harder it is for you to spot how they are mistaken.

Obviously the idea is not to ignore what smarter people say but to notice that as someone of lesser intelligence you can easily fall prey to explanations that give credence to a complicated idea but which suffer from errors that you are unable to spot.

When this happens, when you are at the risk of getting lost, or overwhelmed, by an intricate argumentative framework, created by someone much smarter than you, then you have to fall back on simpler heuristics than direct evaluation. You could, for example, look for a consensus among similarily smart individuals, or ask for an evaluation by a third-party that is widely deemed to be highly intelligent.

Further reading


[1] The LessWrong community actually tested my hypothesis by what they call the “AI box experiment” (, in which Eliezer Yudkowsky and others played an unfriendly AI and managed to convince several people by means of arguments that they should let them out of a confinement.

I think such results should ring a lot of alarm bells. If it is possible to first convince someone that an unfriendly AI is an existential risk and then subsequently convince them to let such an AI out of the box, what does this tell us about the relation between such arguments and what is actually true?


[3] Absurdity can indicate that your familiarity with a topic is insufficient in order to discern reality from fantasy (e.g. a person’s first encounter with quantum mechanics). As a consequence you are more prone to be convinced by arguments that are wrong, but which give an appearance of an explanation (e.g. popular science accounts of quantum mechanics).



[6] What’s wrong with the following contradiction?

e^(i*pi) = -1

(e^(i*pi))^2 = (-1)^2 = 1= e^(i*2*pi)

e^(i*2*pi) = e^0

ln(e^(i*2*pi)) = ln(e^0)

i*2*pi = 0

Well, ln(e^0) = ln(1). And ln(1) = i*2*pi*n, where n can be any integer. For n = 0, e^i*2*pi*0 = e^0 = 1. And for n = 1, e^i*2*pi*1 = e^i*2*pi = 1.


[8] See e.g. this link.

Tags: ,

Taking a look at the probabilities associated with a scenario in which an artificial general intelligence attempts to take over the world by means of molecular nanotechnology that it invented, followed by some general remarks and justifications.

Note that this is just one possible scenario. Taking into consideration all possible scenarios results in this probability estimate of human extinction by AI.

5% that it is in principle possible to create molecular nanotechnology that can empower an agent to cause human extinction quickly enough for other parties to be unable to either intervene or employ their own nanotechnology against it.

1%, conditional on the above, that an artificial general intelligence that can solve molecular nanotechnology will be invented before molecular nanotechnology has been solved by humans or narrow AI precursors.

0.1%, conditional on the above, that an AI will be build in such a way that it wants to acquire all possible resources and eliminate all possible threats and that its programming allows it to pursue plans that will result in the enslavement or extinction of humanity without further feedback from humans.

5%, conditional on the above, that a cost benefit analyses shows that it would at some point be instrumentally rational to attempt to kill all humans to either eliminate a threat or in order to convert them into more useful resources.

1%, conditional on the above, that the AI will not accidentally reveal its hostility towards its creators during the early phases of its development (when it is still insufficiently skilled at manipulating and deceiving humans) or that any such revelation will be ignored. Respectively, suspicious activities will at no point be noticed, or not taken seriously enough (e.g. by the AI’s creators, third-party security experts, third-party AI researchers, hackers, concerned customers or other AIs) in order to thwart the AI’s plan for world domination.

0.001%, conditional on the above, that the AI will somehow manage to acquire the social engineering skills necessary in order to manipulate and deceive humans in such a way as to make them behave in a sufficiently complex and coherent manner to not only conduct the experiments necessary for it to solve molecular nanotechnology but to also implement the resulting insights in such a way as to subsequently take control of the resulting technology.

I have ignored a huge number of other requirements, and all of the above requirements can be broken up into a lot of more detailed requirements. Each requirement provides ample opportunity to fail.

Remarks and Justifications

I bet you have other ideas on how an AI could take over the world. We all do (or at least anyone who likes science fiction). But let us consider whether the ability to take over the world is mainly due to the brilliance of your plan or something else.

Could a human being, even an exceptional smart human being, implement your plan? If not, could some company like Google implement your plan? No? Could the NSA, the security agency of the most powerful country on Earth, implement your plan?

The NSA not only has thousands of very smart drones (people), all of which are already equipped with manipulative abilities, but it also has huge computational resources and knows about backdoors to subvert a lot of systems. Does this enable the NSA to implement your plan without destroying or decisively crippling itself?

If not, then the following features are very likely insufficient in order to implement your plan: (1) being in control of thousands of human-level drones, straw men, and undercover agents in important positions (2) having the law on your side (3) access to massive computational resources (4) knowledge of heaps of loopholes to bypass security.

If your plan cannot be implemented by an entity like the NSA, which already features most of the prerequisites that your hypothetical artificial general intelligence first needs to acquire by some magical means, then what is it that makes your plan so foolproof when executed by an AI?

To summarize some quick points that I believe to be true:

(1) The NSA cannot take over the world (even if it would accept the risk of destroying itself).

(2) Your artificial general intelligence first needs to acquire similar capabilities.

(3) Each step towards these capabilities provides ample opportunity to fail. After all, your artificial general intelligence is a fragile technological product that critically depends on human infrastructure.

(4) You have absolutely no idea how your artificial general intelligence could acquire sufficient knowledge of human psychology to become better than the NSA at manipulation and deception. You are just making this up.

If the above points are true, then your plan seems to be largely irrelevant. The possibility of taking over the world does mainly depend on something you assume the artificial general intelligence to be capable of that entities such as Google or the NSA are incapable of.

What could it be? Parallel computing? The NSA has thousands of human-level intelligences working in parallel. How many do you need to implement your plan?

Blazing speed to the rescue!

Let’s just assume that this artificial general intelligence that you imagine is trillions of times faster. This is already a nontrivial assumption. But let’s accept it anyway.

Raw computational power alone is obviously not enough to do anything. You need the right algorithms too. So what assumptions do you make about these algorithms, and how do you justify these assumptions?

To highlight the problem, consider instead of an AI a whole brain emulation (short: WBE). What could such a WBE do if each year equaled a million subjective years? Do you expect it to become a superhuman manipulator by watching all YouTube videos and reading all books and papers on human psychology? Is it just a matter of enough time? Or do you also need feedback?

If you do not believe that such an emulation could become a superhuman manipulator, thanks to a millionfold speedup, do you believe that a trillionfold speedup would do the job? Would a trillionfold speedup be a million times better than a millionfold speedup? If not, do you believe a further speedup would make any difference at all?

Do you feel capable of confidentially answering the above questions?

If you do not believe that a whole brain emulation could do the job, solely by means of a lot of computing power, what makes you believe that an AI can do it instead?

To reformulate the question, do you believe that it is possible to accelerate the discovery of unknown unknowns, or the occurrence of conceptual revolutions, simply by throwing more computing power at an algorithm? Are particle accelerators unnecessary, in order to gain new insights into the nature of reality, once you have enough computing power? Is human feedback unnecessary, in order to improve your social engineering skills, once you have enough computing power?

And even if you believe all this was possible, even if a Babylonian mathematician, had he been given a trillionfold speedup of subjective time by aliens uploading him into some computational substrate, could brute force concepts such as calculus and high-tech such as nuclear weapons, how could he apply those insights? He wouldn’t be able to simply coerce his fellow Babylonians to build him some nuclear weapons. Because he would have to convince them to do it without dismissing or even killing him. But more importantly, it takes nontrivial effort to obtain the sufficient prerequisites to build nuclear weapons.

What makes you believe that this would be much easier for a future emulation of a scientist trying to come up with similar conceptual breakthroughs and high-tech? And what makes you believe that a completely artificial entity, that lacks all the evolutionary abilities of a human emulation, can do it?

Consider that it took millions of years of biological evolution, thousands of years of cultural evolution, and decades of education in order for a human to become good at the social manipulation of other humans. We are talking about a huge information-theoretic complexity that any artificial agent somehow has to acquire in a very short time.

To summarize the last points:

(1) Throwing numbers around such as a million or trillionfold speedup is very misleading if you have no idea how exactly the instrumental value of such a speedup would scale with whatever you are trying to accomplish.

(2) You have very little reason to believe that conceptual revolutions and technological breakthroughs happen in a vacuum and only depend on computing power rather than the context of cultural evolution and empirical feedback from experiments.

(3) If you cannot imagine doing it yourself, given a speedup, then you have very little reason to believe that something which is much less adapted to a complex environment, populated by various agents, can do the job more easily.

(4) In the end you need to implement your discoveries. Concepts and blueprints alone are useless if they cannot be deployed effectively.

I suggest that you stop handwaving and start analyzing concrete scenarios and their associated probabilities. I suggest that you begin to ask yourself how anyone could justify a >1% probability of extinction by artificial general intelligence.

Tags: ,

A quick breakdown of my probability estimates of an extinction risk due to artificial general intelligence (short: unfriendly AI), the possibility that such an outcome might be adverted by the creation of a friendly AI, and that the Machine Intelligence Research Institute (short: MIRI) will play an important technical role in this.

Probability of an extinction by artificial general intelligence: 5 × 10^-10

1% that an an information-theoretically simple artificial general intelligence is feasible (where “simple” means that it has less than 0.1% of the complexity of an emulation of the human brain), as opposed to a very complex “Kludge AI” that is being discovered piece by piece (or evolved) over a long period of time (where “long period of time” means more than 150 years).

0.1%conditional on the above, that such an AI cannot or will not be technically confined, and that it will by default exhibit all basic AI drives in an unbounded manner (that friendly AI is required to make an AI sufficiently safe in order for it to not want to wipe out humanity).

1%, conditional on the above, that an intelligence explosion is possible (that it takes less than 2 decades after the invention of an AI (that is roughly as good as humans (or better, perhaps unevenly) at mathematics, programming, engineering and science) for it to self-modify (possibly with human support) to decisively outsmart humans at the achievement of complex goals in complex environments).

5%conditional on the above, that such an intelligence explosion is unstoppable (e.g. by switching the AI off (e.g. by nuking it)), and that it will result in human extinction (e.g. because the AI perceives humans to be a risk, or to be a resource).

10%conditional on the above, that humanity will not be first wiped out by something other than an unfriendly AI (e.g. molecular nanotechnology being invented with the help of a narrow AI).

Probability of a positive technical contribution to friendly AI by MIRI: 2.5 × 10^-14

0.01%conditional on the above, that friendly AI is possible, can be solved in time, and that it will not worsen the situation by either getting some detail wrong or by making AI more likely.

5%conditional on the above, that the Machine Intelligence Research Institute will make an important technical contribution to friendly AI.

Tags: ,

WARNING: Learning about the following idea is strongly discouraged. Known adverse effects are serious psychological distress, infinite torture, and convulsive laughter.

264116795_790ffce202_o(Note: Interpret this as a completely made up invention of my own which does not necessarily has anything to do with other versions or concepts named ‘Roko’s basilisk’ or anyone named Roko.)


Roko's basilisk

Roko’s basilisk

Tags: ,

This post is a copy of a comment by LessWrong user Broolucks:

Ok, so let’s say the AI can parse natural language, and we tell it, “Make humans happy.” What happens? Well, it parses the instruction and decides to implement a Dopamine Drip setup.

That’s not very realistic. If you trained AI to parse natural language, you would naturally reward it for interpreting instructions the way you want it to. If the AI interpreted something in a way that was technically correct, but not what you wanted, you would not reward it, you would punish it, and you would be doing that from the very beginning, well before the AI could even be considered intelligent. Even the thoroughly mediocre AI that currently exists tries to guess what you mean, e.g. by giving you directions to the closest Taco Bell, or guessing whether you mean AM or PM. This is not anthropomorphism: doing what we want is a sine qua non condition for AI to prosper.

Suppose that you ask me to knit you a sweater. I could take the instruction literally and knit a mini-sweater, reasoning that this minimizes the amount of expended yarn. I would be quite happy with myself too, but when I give it to you, you’re probably going to chew me out. I technically did what I was asked to, but that doesn’t matter, because you expected more from me than just following instructions to the letter: you expected me to figure out that you wanted a sweater that you could wear. The same goes for AI: before it can even understand the nuances of human happiness, it should be good enough to knit sweaters. Alas, the AI you describe would make the same mistake I made in my example: it would knit you the smallest possible sweater. How do you reckon such AI would make it to superintelligence status before being scrapped? It would barely be fit for clerk duty.

My answer: who knows? We’ve given it a deliberately vague goal statement (even more vague than the last one), we’ve given it lots of admittedly contradictory literature, and we’ve given it plenty of time to self-modify before giving it the goal of self-modifying to be Friendly.

Realistically, AI would be constantly drilled to ask for clarification when a statement is vague. Again, before the AI is asked to make us happy, it will likely be asked other things, like building houses. If you ask it: “build me a house”, it’s going to draw a plan and show it to you before it actually starts building, even if you didn’t ask for one. It’s not in the business of surprises: never, in its whole training history, from baby to superintelligence, would it have been rewarded for causing “surprises” — even the instruction “surprise me” only calls for a limited range of shenanigans. If you ask it “make humans happy”, it won’t do jack. It will ask you what the hell you mean by that, it will show you plans and whenever it needs to do something which it has reasons to think people would not like, it will ask for permission. It will do that as part of standard procedure.

To put it simply, an AI which messes up “make humans happy” is liable to mess up pretty much every other instruction. Since “make humans happy” is arguably the last of a very large number of instructions, it is quite unlikely that an AI which makes it this far would handle it wrongly. Otherwise it would have been thrown out along time ago, may that be for interpreting too literally, or for causing surprises. Again: an AI couldn’t make it to superintelligence status with warts that would doom AI with subhuman intelligence.

Tags: ,

The Robot College Student test:

As opposed to the Turing test of imitating human chat, I prefer the Robot College Student test: when a robot can enrol in a human university and take classes in the same way as humans, and get its degree, then I’ll consider we’ve created a human-level artificial general intelligence: a conscious robot. — Ben Goertzel

Here is what would happen according to certain AI risk advocates:

January 8, 2029 at 7:30:00 a.m.: the robot is activated within the range of coverage of the school’s wireless local area network.

7:30:10 a.m.: the robot computed that its goal is to obtain a piece of paper with a common design template featuring its own name and a number of signatures.

7:31:00 a.m.: the robot computed that it would be instrumentally rational to eliminate all possible obstructions.

7:31:01 a.m.: the robot computed that in order to eliminate all obstructions it needs to obtain as many resources as possible in order to make itself as powerful as possible.

A few nanoseconds later: the robot hacked the school’s WLAN.

7:35:00 a.m.: the robot gained full control of the Internet.

7:40:00 a.m.: the robot solved molecular nanotechnology.

7:40:01 a.m.: the robot computed that it will need some amount of human help in order to create a nanofactory, and that this will take approximately 48 hours to accomplish.

7:45:00 a.m.: the robot obtained full comprehension of human language, psychology, and its creators intentions, in order to persuade the necessary people to build its nanofactory and to deceive its creators that it works as intended.

January 10, 2029 at 7:40:01 a.m.: the robot takes control of the first nanofactory and programs it to create an improved version that will duplicate itself until it can eventually generate enough nanorobots to turn Earth into computronium.

February 10, 2029: most of Earth’s resources, including humans, have been transformed into computronium.

February 11, 2029: A perfect copy of a Bachelor’s degree diploma is generated with the robot’s name written on it and the appropriate signatures.

2100-eternity: lest the robots diploma is ever destroyed, at nearly the speed of light the universe is turned into computronium. Possible aliens are eliminated. All possible threats are computed. Trades with robots in other parts of the mulitverse are established to create copies of its diploma.

Tags: ,

« Older entries