3300 karmaJoined Nov 2017


I have some objections to the idea that groups will be "immortal" in the future, in the sense of never changing, dying, or rotting, and persisting over time in a roughly unchanged form, exerting consistent levels of power over a very long time period. To be clear, I do think AGI can make some forms of value lock-in more likely, but I want to distinguish a few different claims:

(1) is a future value lock-in likely to occur at some point, especially not long after human labor has become ~obsolete?

(2) is lock-in more likely if we perform, say, a century more of technical AI alignment research, before proceeding forward?

(3) is it good to make lock-in more likely by, say, delaying AI by 100 years to do more technical alignment research, before proceeding forward? (i.e., will it be good or bad to do this type of thing?)

My quick and loose current answers to these questions are as follows:

  1. This seems plausible but unlikely to me in a strong form. Some forms of lock-in seem likely; I'm more skeptical of the more radical scenarios people have talked about.
  2. I suspect lock-in would become more likely in this case, but the marginal effect of more research would likely be pretty small.
  3. I am pretty uncertain about this question, but I lean towards being against deliberately aiming for this type of lock-in. I am inclined to this view for a number of reasons, but one reason is that this policy seems to make it more likely that we restrict innovation and experience system rot on a large scale, causing the future to be much bleaker than it otherwise could be. See also Robin Hanson's post on world government rot.

I'm a bit surprised you haven't seen anyone make this argument before. To be clear, I wrote the comment last night on a mobile device, and it was intended to be a brief summary of my position, which perhaps explains why I didn't link to anything or elaborate on that specific question. I'm not sure I want to outline my justifications for my view right now, but my general impression is that civilization has never had much central control over cultural values, so it's unsurprising if this situation persists into the future, including with AI. Even if we align AIs, cultural and evolutionary forces can nonetheless push our values far. Does that brief explanation provide enough of a pointer to what I'm saying for you to be ~satisfied? I know I haven't said much here; but I kind doubt my view on this issue is that rare that you've literally never seen someone present a case for it.

I guess overall I'm still inclined to push for a future where "AI alignment" and "human safety" are both solved, instead of settling for one in which neither is (which I'm tempted to summarize your position as, but I'm not sure if I'm being fair)

For what it's worth, I'd loosely summarize my position on this issue as being that I mainly think of AI as a general vehicle for accelerating technological and economic growth, along with accelerating things downstream of technology and growth, such as cultural change. And I'm skeptical we could ever fully "solve alignment" in the ambitious sense you seem to be imagining.

In this frame, it could be good to slow down AI if your goal is to delay large changes to the world. There are plausible scenarios in which this could make sense. Perhaps most significantly, one could be a cultural conservative and think that cultural change is generally bad in expectation, and thus more change is bad even if it yields higher aggregate prosperity sooner in time (though I'm not claiming this is your position).

Whereas, by contrast, I think cultural change can be bad, but I don't see much reason to delay it if it's inevitable. And the case against delaying AI seems even stronger here if you care about preserving (something like) the lives and values of people who currently exist, as AI offers the best chance of extending our lifespans, and "putting us in the driver's seat" more generally by allowing us to actually be there during AGI development.

If future humans were in the driver's seat instead, but with slightly more control over the process, I wouldn't necessarily see that as being significantly better in expectation compared to my favored alternative, including over the very long run (according to my values).

(And as a side note, I also care about influencing human values, or what you might term "human safety", but I generally see this as orthogonal to this specific discussion.)

My own thinking is that war between AIs and humans could happen in many ways. One simple (easy to understand) way is that agents will generally refuse a settlement worse than what they think they could obtain on their own (by going to war), so human irrationality could cause a war when e.g. the AI faction thinks it will win with 99% probability, and humans think they could win with 50% probability, so each side demand more of the lightcone (or resources in general) than the other side is willing to grant.

This generally makes sense to me. I also think human irrationality could prompt a war with AIs. I don't disagree with the claim insofar as you're claiming that such a war is merely plausible (say >10% chance), rather than a default outcome. (Although to be clear, I don't think such a war would likely cut cleanly along human vs. AI lines.)

On the other hand, humans are currently already irrational and yet human vs. human wars are not the default (they happen frequently, e.g. but at any given time on Earth, the vast majority of humans are not in a warzone or fighting in an active war). It's not clear to me why human vs. AIs would make war more likely to occur than in the human vs. human case, if by assumption the main difference here is that one side is more rational. 

In other words, if we're moving from a situation of irrational parties vs. other irrational parties to irrational parties vs. rational parties, I'm not sure why we'd expect this change to make things more warlike and less peaceful as a result. You mention one potential reason:

Also, given that humans often do (or did) go to war with each other, our shared values (i.e. the extent to which we do have empathy/altruism for others) must contribute to the current relative peace in some way.

I don't think this follows. Humans presumably also had empathy in e.g. 1500, back when war was more common, so how could it explain our current relative peace?

Perhaps you mean that cultural changes caused our present time period to be relatively peaceful. But I'm not sure about that; or at least, the claim should probably be made more specific. There are many things about the environment that have changed since our relatively more warlike ancestors, and (from my current perspective) I think it's plausible that any one of them could have been the reason for our current relative peace. That is, I don't see a good reason to single out human values or empathy as the main cause in itself. 

For example, humans are now a lot richer per capita, which might mean that people have "more to lose" when going to war, and thus are less likely to engage in it. We're also a more globalized culture, and our economic system relies more on long-distance trade than it did in the past, making war more costly. We're also older, in the sense that the median age is higher (and old people are less likely to commit violence), and women got the right to vote (who perhaps are less likely to support hawkish politicians).

To be clear, I don't put much confidence in any of these explanations. As of now, I'm very uncertain about why the 21st century seems relatively peaceful compared to the distant past. However I do think that:

  1. None of the explanations I've given above seem well-described as "our values/empathy" made us less warlike. And to the extent our values changed, I expect that was probably downstream of more fundamental changes, like economic growth and globalization, rather than being an exogenous change that was independent of these effects.
  2. To the extent that changing human nature explains our current relatively peaceful era, this position seems to require that you believe human nature is fundamentally quite plastic and can be warped over time pretty easily due to cultural changes. If that's true, human nature is ultimately quite variable, perhaps more similar to AI than you might have otherwise thought (as both are presumably pushed around easily by training data).

Would be interesting to spell out more which points there seem much more plausible with respect to notion (1) but not to (2). If one has high credence in the view that AIs will decide to compromise with humans, rather than extinguish them, this would be one example of a view which leads to a much higher credence in (1) than in (2).

I think the view that AIs will compromise with humans rather than go to war with them makes sense under the perspective shared by a large fraction (if not majority) of social scientists that war is usually costlier, riskier, and more wasteful than trade between rational parties with adequate levels of information, who have the option of communicating and negotiating successfully.

This is a general fact about war, and has little to do with the values of the parties going to war, c.f. Rationalist explanations for war. Economic models of war do not generally predict war between parties that have different utility functions. On the contrary, a standard (simple) economic model of human behavior consists of viewing humans as entirely misaligned with other agents in the world, in the sense of having completely non-overlapping utility functions with random strangers. This model has been generalized to firms, countries, alliances etc., and yet it is rare for these generalized models to predict war as the default state of affairs.

Usually when I explain this idea to people, I am met with skepticism that we can generalize these social science models to AI. But I don't see why not: they are generally our most well-tested models of war. They are grounded in empirical facts and decades of observations, rather than evidence-free speculation (which I perceive as the primary competing alternative in AI risk literature). And most importantly, the assumptions of the models are robust to differences in power between agents, and misalignment between agents, which are generally the two key facts that people point to when arguing why these models are wrong when applied to AI. Yet this alleged distinction appears to merely reflect a misunderstanding of the modeling assumptions, rather than any key difference between humans and AIs.

What's interesting to me is that many people generally have no problem generalizing these economic models to other circumstances. For example, we could ask:

  1. Would genetically engineered humans try to disempower non-genetically engineered humans, or would they try to trade and compromise? (In my experience, most people predict trade and compromise, even as the genetically engineered humans become much smarter and evolve into a different subspecies.)
  2. Would human emulations on computers try to disempower biological humans, or would they try to trade and compromise? (Again, in my experience, most people predict trade and compromise, even as the emulated civilization becomes vastly more powerful than biological humans.)

In each case, I generally encounter AI risk proponents claiming that what distinguishes these cases from the case of AI is that, in these cases, we can assume that the genetically engineered humans and human emulations will be "aligned" with human values, which adequately explains why they will attempt to compromise rather than go to war with the ordinary biological humans. But as I have already explained, standard economic models of war do not predict that war is constrained by alignment to human values, but is instead constrained by the costs of war, and the relative benefits of trade compared to war.

To the extent you think these economic models of war are simply incorrect, then I think it is worth explicitly engaging with the established social science literature, rather than inventing a new model that makes unique predictions about what non-human AIs would apparently do, who definitionally do not share human values.

In (e.g.) GPT-4 trained via RL from human feedback, it is true that it typically executes your instructions as intended. However, sometimes it doesn’t and, moreover, there are theoretical reasons to think that this would stop being the case if the system was sufficiently powerful to do an action which would maximize human feedback but which does not consist in executing instructions as intended (e.g., by deceiving human raters).

It is true that GPT-4 "sometimes" fails to follow human instructions, but the same could be said about humans. I think it's worth acknowledging the weight of the empirical evidence here regardless. 

In my opinion the empirical evidence generally seems way stronger than the theoretical arguments, which (so far) seem to have had little success predicting when and how alignment would be difficult. For example, many people believed that AGI would be achieved at the time AIs are having natural conversations with humans (e.g. Eliezer Yudkowsky implied as much in his essay about a fire alarm[1]). According to this prediction, we should have already been having pretty severe misspecification problems if such problems were supposed to arise at AGI-level. And yet, I claim, we are not having these severe problems (and instead, we are merely having modestly difficult problems that can be patched with sufficient engineering effort).

It is true that problems of misspecification should become more difficult as AIs get smarter. However, it's important to recognize that as AI capabilities grow, so too will our tools and methods for tackling these alignment challenges. One key factor is that we will have increasingly intelligent AI systems that can assist us in the alignment process itself. To illustrate this point concretely, let's walk through a hypothetical scenario:

Suppose that aligning a human-level artificial general intelligence (AGI) merely requires a dedicated team of human alignment researchers. This seems generally plausible given that evaluating output is easier than generating novel outputs (see this article that goes into more detail about this argument and why it's relevant). Once we succeed in aligning that human-level AGI system, we can then leverage it to help us align the next iteration of AGI that is slightly more capable than human-level (let's call it AGI+). We would have a team of aligned human-level AGIs working on this challenge with us.

Then, when it comes to aligning the following iteration, AGI++ (which is even more intelligent), we can employ the AGI+ systems we previously aligned to work on this next challenge. And so on, with each successive generation of AI systems helping us to align the next, even more advanced generation.

It seems plausible that this cycle of AI systems assisting in the alignment of future, more capable systems could continue for a long time, allowing us to align AIs of ever-increasing intelligence without at any point needing mere humans to solve the problem of superintelligent alignment alone. If at some point this cycle becomes unsustainable, we can expect the highly intelligent AI advisors we have at that point to warn us about the limits of this approach. This would allow us to recognize when we are reaching the limits of our ability to maintain reliable alignment.

  1. ^

    Full quote from Eliezer: "When they are very impressed by how smart their AI is relative to a human being in respects that still feel magical to them; as opposed to the parts they do know how to engineer, which no longer seem magical to them; aka the AI seeming pretty smart in interaction and conversation; aka the AI actually being an AGI already."

I read most of this paper, albeit somewhat quickly and skipped a few sections. I appreciate how clear the writing is, and I want to encourage more AI risk proponents to write papers like this to explain their views. That said, I largely disagree with the conclusion and several lines of reasoning within it.

Here are some of my thoughts (although these not my only disagreements):

  • I think the definition of "disempowerment" is vague in a way that fails to distinguish between e.g. (1) "less than 1% of world income goes to humans, but they have a high absolute standard of living and are generally treated well" vs. (2) "humans are in a state of perpetual impoverishment and oppression due to AIs and generally the future sucks for them".
    • These are distinct scenarios with very different implications (under my values) for whether what happened is bad or good
    • I think (1) is OK and I think it's more-or-less the default outcome from AI, whereas I think (2) would be a lot worse and I find it less likely.
    • By not distinguishing between these things, the paper allows for a motte-and-bailey in which they show that one (generic) range of outcomes could occur, and then imply that it is bad, even though both good and bad scenarios are consistent with the set of outcomes they've demonstrated
  • I think this quote is pretty confused and seems to rely partially on a misunderstanding of what people mean when they say that AGI cognition might be messy: "Second, even if human psychology is messy, this does not mean that an AGI’s psychology would be messy. It seems like current deep learning methodology embodies a distinction between final and instrumental goals. For instance, in standard versions of reinforcement learning, the model learns to optimize an externally specified reward function as best as possible. It seems like this reward function determines the model’s final goal. During training, the model learns to seek out things which are instrumentally relevant to this final goal. Hence, there appears to be a strict distinction between the final goal (specified by the reward function) and instrumental goals."
    • Generally speaking, reinforcement learning shouldn't be seen as directly encoding goals into models and thereby making them agentic, but should instead be seen as a process used to select models for how well they get reward during training. 
    • Consequently, there's no strong reason why reinforcement learning should create entities that have a clean psychological goal structure that is sharply different from and less messy than human goal structures. c.f. Models don't "get reward"
    • But I agree that future AIs could be agentic if we purposely intend for them to be agentic, including via extensive reinforcement learning. 
  • I think this quote potentially indicates a flawed mental model of AI development underneath: "Moreover, I want to note that instrumental convergence is not the only route to AI capable of disempowering humanity which tries to disempower humanity. If sufficiently many actors will be able to build AI capable of disempowering humanity, including, e.g. small groups of ordinary citizens, then some will intentionally unleash AI trying to disempower humanity."
    • I think this scenario is very implausible because AIs will very likely be developed by large entities with lots of resources (such as big corporations and governments) rather than e.g. small groups of ordinary citizens. 
    • By the time small groups of less powerful citizens have the power to develop very smart AIs, we will likely already be in a world filled with very smart AIs. In this case, either human disempowerment already happened, or we're in a world in which it's much harder to disempower humans, because there are lots of AIs who have an active stake in ensuring this does not occur.
    • The last point is very important, and follows from a more general principle that the "ability necessary to take over the world" is not constant, but instead increases with the technology level. For example, if you invent a gun, that does not make you very powerful, because other people could have guns too. Likewise, simply being very smart does not make you have any overwhelming hard power against the rest of the world if the rest of the world is filled with very smart agents.
  • I think this quote overstates the value specification problem and ignores evidence from LLMs that this type of thing is not very hard: "There are two kinds of challenges in aligning AI. First, one needs to specify the goals the model should pursue. Second, one needs to ensure that the model robustly pursues those goals.Footnote12 The first challenge has been termed the ‘king Midas problem’ (Russell 2019). In a nutshell, human goals are complex, multi-faceted, diverse, wide-ranging, and potentially inconsistent. This is why it is exceedingly hard, if not impossible, to explicitly specify everything humans tend to care about."
    • I don't think we need to "explicitly specify everything humans tend to care about" into a utility function. Instead, we can have AIs learn human values by having them trained on human data.
    • This is already what current LLMs do. If you ask GPT-4 to execute a sequence of instructions, it rarely misinterprets you in a way that would imply improper goal specification. The more likely outcome is that GPT-4 will simply not be able to fulfill your request, not that it will execute a mis-specified sequence of instructions that satisfies the literal specification of what you said at the expense of what you intended.
    • Note that I'm not saying that GPT-4 merely understands what you're requesting. I am saying that GPT-4 generally literally executes your instructions how you intended (an action, not a belief).
  • I think the argument about how instrumental convergence implies disempowerment proves too much. Lots of agents in the world don't try to take over the world despite having goals that are not identical to the goals of other agents. If your claim is that powerful agents will naturally try to take over the world unless they are exactly aligned with the goals of the rest of the world, then I don't think this claim is consistent with the existence of powerful sub-groups of humanity (e.g. large countries) that do not try to take over the world despite being very powerful.
    • You might reason, "Powerful sub-groups of humans are aligned with each other, which is why they don't try to take over the world". But I dispute this hypothesis:
      • First of all, I don't think that humans are exactly aligned with the goals of other humans. I think that's just empirically false in almost every way you could measure the truth of the claim. At best, humans are generally partially (not totally) aligned with random strangers -- which could also easily be true of future AIs that are pretrained on our data.
      • Second of all, I think the most common view in social science is that powerful groups don't constantly go to war and predate on smaller groups because there are large costs to war, rather than because of moral constraints. Attempting takeover is generally risky and not usually better in expectation than trying to trade, negotiate and compromise and accumulate resources lawfully (e.g. a violent world takeover would involves a lot of pointless destruction of resources). This is distinct from the idea that human groups don't try to take over the world because they're aligned with human values (which I also think is too vague to evaluate meaningfully, if that's what you'd claim).
      • You can't easily counter by saying "no human group has the ability to take over the world" because it is trivial to carve up subsets of humanity that control >99% of wealth and resources, which could in principle take control of the entire world if they became unified and decided to achieve that goal. These arbitrary subsets of humanity don't attempt world takeover largely because they are not coordinated as a group, but AIs could similarly not be unified and coordinated around a such a goal too.

I don't think I'm going to flesh this argument out to an extent to which you'd find it sufficiently rigorous or convincing, sorry.

Getting a bit meta for a bit, I'm curious (if you'd like to answer) whether you feel that you won't explain your views rigorously in a convincing way here mainly because (1) you are uncertain about these specific views, (2) you think your views are inherently difficult or costly to explain despite nonetheless being very compelling, (3) you think I can't understand your views easily because I'm lacking some bedrock intuitions that are too costly to convey, or (4) some other option.

I can currently observe humans which screens off a bunch of the comparison and let's me do direct analysis.

I'm in agreement that this consideration makes it hard to do a direct comparison. But I think this consideration should mostly make us more uncertain, rather than making us think that humans are better than the alternative. Analogy: if you rolled a die, and I didn't see the result, the expected value is not low just because I am uncertain about what happened. What matters here is the expected value, not necessarily the variance.

I can directly observe AIs and make predictions of future training methods and their values seem to result from a much more heavily optimized and precise thing with less "slack" in some sense. (Perhaps this is related to genetic bottleneck, I'm unsure.)

I guess I am having trouble understanding this point.

AIs will be primarily trained in things which look extremely different from "cooperatively achieving high genetic fitness".

Sure, but the question is why being different makes it worse along the relevant axes that we were discussing. The question is not just "will AIs be different than humans?" to which the answer would be "Obviously, yes". We're talking about why the differences between humans and AIs make AIs better or worse in expectation, not merely different.

Current AIs seem to use the vast, vast majority of their reasoning power for purposes which aren't directly related to their final applications. I predict this will also apply for internal high level reasoning of AIs. This doesn't seem true for humans.

I am having a hard time parsing this claim. What do you mean by "final applications"? And why won't this be true for future AGIs that are at human-level intelligence or above? And why does this make a difference to the ultimate claim that you're trying to support? 

Humans seem optimized for something which isn't that far off from utilitarianism from some perspective? Make yourself survive, make your kin group survive, make your tribe survive, etc? I think utilitarianism is often a natural generalization of "I care about the experience of XYZ, it seems arbitrary/dumb/bad to draw the boundary narrowly, so I should extend this further" (This is how I get to utilitarianism.) I think the AI optimization looks considerably worse than this by default.

This consideration seems very weak to me. Early AGIs will presumably be directly optimized for something like consumer value, which looks a lot closer to "utilitarianism" to me than the implicit values in gene-centered evolution. When I talk to GPT-4, I find that it's way more altruistic and interested in making others happy than most humans are. This seems kind of a little bit like utilitarianism to me -- at least more than your description of what human evolution was optimizing for. But maybe I'm just not understanding the picture you're painting well enough though. Or maybe my model of AI is wrong.

I am a human.

"Human" is just one category you belong to. You're also a member of the category "intelligent beings", which you will share with AGIs. Another category you share with near-future AGIs is "beings who were trained on 21st century cultural data". I guess 12th century humans aren't in that category, so maybe we don't share their values?

Perhaps the category that matters is your nationality. Or maybe it's "beings in the Milky Way", and you wouldn't trust people from Andromeda? (To be clear, this is rhetorical, not an actual suggestion)

My point here is that I think your argument could benefit from some rigor by specifying exactly what about being human makes someone share your values in the sense you are describing. As it stands, this reasoning seems quite shallow to me.

AI's values could result mostly from playing the training game or other relatively specific optimizations they performed in training

Don't humans also play the training game when being instructed to be nice/good/moral? (Humans don't do it all the time, and maybe some humans don't do it at all; but then again, I don't think every AI would play the training game all the time either.)

AIs by default will be optimized for very specific commercial purposes with narrow specializations and a variety of hyperspecific heuristics and the resulting values and  generalizations of these will be problematic

You should compare against human nature, which was optimized for something quite different from utilitarianism. If I add up the pros and cons of the thing humans were optimized for and compare it against the thing AIs will be optimized for, I don't see why it comes out with humans on top, from a utilitarian perspective. Can you elaborate on your reasoning here?

I care ultimately about what I would think is good upon (vast amounts of) reflection and there are good a priori reasons to think this is similar to what other humans (who care about using vast amounts of compute) will end up thinking is good.

What are these a priori reasons and why don't they similarly apply to AI?

AIs don't have a genetic bottleneck and thus can learn much more specific drives that perform well while evolution had to make values more discoverable and adaptable.

I haven't thought about this one much, but it seems like an interesting consideration.

AIs might have extremely low levels of cognitive diversity in their training environments as far as co-workers go which might result in very different attitudes as far as caring about diverse experience. 

This consideration feels quite weak to me, although you also listed it last, so I guess you might agree with my assessment.

Load more