Jessica Taylor. CS undergrad and Master’s at Stanford; former research fellow at MIRI.
I work on decision theory, social epistemology, strategy, naturalized agency, mathematical foundations, decentralized networking systems and applications, theory of mind, and functional programming languages.
Blog: unstableontology.com
Twitter: https://twitter.com/jessi_cata
Computationally tractable is Yudkowsky’s framing and might be too limited. The kind of thing I believe is for example, an animal without a certain brain complexity will tend not to be a social animal and is therefore unlikely to have the sort of values social animals have. And animals that can’t do math aren’t going to value mathematical aesthetics the way human mathematicians do.
Relativity to Newtonian mechanics is a warp in a straightforward sense. If you believe the layout of a house consists of some rooms connected in a certain way, but there are actually more rooms connected in different ways, getting the maps to line up looks like a warp. Basically, the closer the mapping is to a true homomorphism (in the universal algebra sense), the less warping there is, otherwise there are deviations intuitively analogous to space warps.
I discussed something similar in the “Human brains don’t seem to neatly factorize” section of the Obliqueness post. I think this implies that, even assuming the Orthogonality Thesis, humans don’t have values that are orthogonal to human intelligence (they’d need to not respond to learning/reflection to be orthogonal in this fashion), so there’s not a straightforward way to align ASI with human values by plugging in human values to more intelligence.
hmm, I wouldn’t think of industrialism and human empowerment as trying to grab the whole future, just part of it, in line with the relatively short term (human not cosmic timescale) needs of the self and extended community; industrialism seems to lead to capitalist organization which leads to decentralization superseding nations and such (as Land argues).
I think communism isn’t generally about having one and one’s friends in charge, it is about having human laborers in charge. One could argue that it tended towards nationalism (e.g. USSR), but I’m not convinced that global communism (Trotskyism) would have worked out well either. Also, one could take an update from communism about agendas for global human control leading to national control (see also tendency of AI safety to be taken over by AI national security as with the Situational Awareness paper). (Again, not ruling out that grabbing hold of the entire future could be a good idea at some point, just not sold on current agendas and wanted to note there are downsides that push against Pascal’s mugging type considerations)
Not sure what you mean by complexity here, is this like code size / Kolmogorov complexity? You need some of that to have intelligence at all (the empty program is not intelligent). At some point most of your gains come from compute rather than code size. Though code size can speed things up (e.g. imagine sending a book back to 1000BC, that would speed people up a lot; consider that superintelligence sending us a book would be a bigger speedup)
by “complexify” here it seems you mean something like “develop extended functional organization”, e.g. in brain development throughout evolution. And yeah, that involves dynamics with the environment and internal maintenance (evolution gets feedback from the environment). It seems it has to have a drive to do this which can either be a terminal or instrumental goal, though deriving it from instrumentals seems harder than baking it is as terminal (so I would guess evolution gives animals a terminal goal of developing functional complexity of mental structures etc, or some other drive that isn’t exactly a terminal goal)
see also my post relating optimization daemons to immune systems, it seems evolved organisms develop these; when having more extended functional organization, they protect it with some immune system functional organization.
to be competitive agents, having a “self” seems basically helpful, but might not be the best solution; selfish genes are an alternative, and perhaps extended notions of self can maintain competitiveness.
Thanks, going to link this!
re meta ethical alternatives:
roughly my view
slight change, opens the question of why the deviations? are the “right things to value” not efficient to value in a competitive setting? mostly I’m trying to talk about those things to value that go along with intelligence, so it wouldn’t correspond with a competitive disadvantage in general. so it’s still close enough to my view
roughly Yudkowskian view, main view under which the FAI project even makes sense. I think one can ask basic questions like which changes move towards more rationality on the margin, though such changes would tend to prioritize rationality over preventing value drift. I’m not sure how much there are general facts about how to avoid value drift (it seems like the relevant kind, i.e. value drift as part of becoming more rational/intelligent, only exists from irrational perspectives, in a way dependent on the mind architecture)
minimal CEV-realist view. it really seems up to agents how much they care about their reflected preferences. maybe changing preferences too often leads to money pumps, or something?
basically says “there are irrational and rational agents, rationality doesn’t apply to irrational agents”, seems somewhat how people treat animals (we don’t generally consider uplifting normative with respect to animals)
at this point you’re at something like ecology / evolutionary game theory, it’s a matter of which things tend to survive/reproduce and there aren’t general decision theories that succeed
re human ontological crises: basically agree, I think it’s reasonably similar to what I wrote. roughly my reason for thinking that it’s hard to solve is that the ideal case would be something like a universal algebra homomorphism (where the new ontology actually agrees with the old one but is more detailed), yet historical cases like physics aren’t homomorphic to previous ontologies in this way, so there is some warping necessary. you could try putting a metric on the warping and minimizing it, but, well, why would someone think the metric is any good, it seems more of a preference than a thing rationality applies to. if you think about it and come up with a solution, let me know, of course.
with respect to grabbing hold of the whole future: you can try looking at historical cases of people trying to grab hold of the future and seeing how that went, it’s a mixed bag with mostly negative reputation, indicating there are downsides as well as upsides, it’s not a “safe” conservative view. see also Against Responsibility. I feel like there’s a risk of getting Pascal’s mugged about “maybe grabbing hold of the future is good, you can’t rule it out, so do it”, there are downsides to spending effort that way. like, suppose some Communists thought capitalism would lead to the destruction of human value with high enough probability that instituting global communism is the conservative option, it doesn’t seem like that worked well (even though a lot of people around here would agree that capitalism tends to leads to human value destruction in the long run). particular opportunities for grabbing hold of the future can be net negative and not worth worrying about even if one of them is a good idea in the long run (I’m not ruling that out, just would have to be convinced of specific opportunities).
overall I’d rather focus on first modeling the likely future and looking for plausible degrees of freedom; a general issue with Pascal’s mugging is it might make people overly attached to world models in which they have ~infinite impact (e.g. Christianity, Communism) which means paying too much attention to wrong world models, not updating to more plausible models in which existential-stakes decisions could be comprehended if they exist. and Obliqueness doesn’t rule out existential stakes (since it’s non-Diagonal).
as another point, Popperian science tends to advance by people making falsifiable claims, “you don’t know if that’s true” isn’t really an objection in that context. the pragmatic claim I would make is: I have some Bayesian reason to believe agents do not in general factor into separate Orthogonal and Diagonal components, this claim is somewhat falsifiable (someone could figure out a theory of this invulnerable to optimization daemons etc), I’m going to spend my attention on the branch where I’m right, I’m not going to worry about Pascal’s mugging type considerations for if I’m wrong (as I said, modeling the world first seems like a good general heuristic), people can falsify it eventually if it’s false.
this whole discussion is not really a defense of Orthogonality given that Yudkowsky presented orthogonality as a descriptive world model, not a normative claim, so sticking to the descriptive level in the original post seems valid; it would be a form of bad epistemology to reject a descriptive update (assuming the arguments are any good) because of pragmatic considerations.
“as important as ever”: no, because our potential influence is lower, and the influence isn’t on things shaped like our values, there has to be a translation, and the translation is different from the original.
CEV: while it addresses “extrapolation” it seems broadly based on assuming the extrapolation is ontologically easy, and “our CEV” is an unproblematic object we can talk about (even though it’s not mathematically formalized, any formalization would be subject to doubt, and even if formalized, we need logical uncertainty over it, and logical induction has additional free parameters in the limit). I’m really trying to respond to orthogonality not CEV though.
from a practical perspective: notice that I am not behaving like Eliezer Yudkowsky. I am not saying the Orthogonality Thesis is true and important to ASI, I am instead saying intelligence/values are Oblique and probably nearly Diagonal (though it’s unclear what I mean by “nearly”). I am not saying a project of aligning superintelligence with human values is a priority. I am not taking research approaches that assume a Diagonal/Orthogonal factorization. I left MIRI partially because I didn’t like their security policies (and because I had longer AI timelines), I thought discussion of abstract research ideas was more important. I am not calling for a global AI shutdown so this project (which is in my view confused) can be completed. I am actually against AI regulation on the margin (I don’t have a full argument for this, it’s a political matter at this point).
I think practicality looks more like having near-term preferences related to modest intelligence increases (as with current humans vs humans with neural nets; how do neural nets benefit or harm you, practically? how can you use them to think better and improve your life?), and not expecting your preferences to extend into the distant future with many ontology changes, so don’t worry about grabbing hold of the whole future etc, think about how to reduce value drift while accepting intelligence increases on the margin. This is a bit like CEV except CEV is in a thought experiment instead of reality.
The “Models of ASI should start with realism” bit IS about practicalities, namely, I think focusing on first forecasting absent a strategy of what to do about the future is practical with respect to any possible influence on the far future; practically, I think your attempted jump to practicality (which might be related to philosophical pragmatism) is impractical in this context.
It occurs to me that maybe you mean something like “Our current (non-extrapolated) values are our real values, and maybe it’s impossible to build or become a superintelligence that shares our real values so we’ll have to choose between alignment and superintelligence.” Is this close to your position?
Close. Alignment of already-existing human values with superintelligence is impossible (I think) because of the arguments given. That doesn’t mean humans have no preferences indirectly relating to superintelligence (especially, we have preferences about modest intelligence increases, and there’s some iterative process).
I think that’s what I’m trying to say with orthogonal and diagonal both being wrong. One example of a free choice would be bets on things that are very hard to test or deduce. Then you decide some probability, and if you change the probability too much you get money pumped as with a logical inductor. But of course thinking and learning more will tend to concentrate beliefs more, so it isn’t truly orthogonal. (One could think values but not beliefs are orthogonal, but we both know about Bayes/VNM duality)
I think the relevant implication from the thought experiment is that thinking a bunch about metaethics and so on will in practice change your values; the pill itself is not very realistic, but thinking can make people smarter and will cause value changes. I would agree Land is overconfident (I think orthogonal and diagonal are both wrong models).
I don’t think it’s a given that moral nonrealism is true (therefore inevitably believed by a superintelligence), see my short story.
Morality can mean multiple things. Utilitarian morality is about acting to maximize a fixed goal function, Kantian morality is about alignment between the a posteriori will and possible a priori will, cultural morality is about adherence to a specific method of organizing humans.
Superintelligence would clearly lack human cultural morality, it’s a specific system organizing humans, e.g. with law as a relatively legible branch.
In general humans question more of their previous morality when thinking longer; Peter Singer for example rejects much of normal morality for utilitarian reasons.
ASI could have something analogous to cultural morality but for organizing a different set of agents. E.g. methods of side-taking in game-theoretic conflict that tend to promote cooperation between different ASIs (this becomes more relevant e.g. when an alien ASI is encountered or more speculatively in acausal trade).
Regardless of whether one calls Omohundro drives “moral”, they are convergent goals for ASIs, so the rejection of human morality does not entail lack of very general motives that include understanding the world and using resources such as energy efficiently and so on.
I think both (a) something like moral realism is likely true and (b) the convergent morality for ASIs does not particularly care about humans if ASIs already exist (humans are of course important in the absence of ASIs due to greater intelligence/agency than other entities on Earth).
FAI is a narrow path to ASI that has similar values to what humans would upon reflection. As I have said these are very different from current human values due to more thought and coherence and so on. It might still disassemble humans but scan them into simulation and augment them, etc. (This is an example of what I referred to as “luxury consumerism in the far future”)
To the extent will-to-think generates a “should” for humans the main one is “you should think about things including what is valuable, and trust the values upon reflection more than current values, rather than being scared of losing current values on account of thinking more”. It’s basically an option for people to do this or not, but as Land suggests, not doing this leads to a competitive disadvantage in the long run. And general “should”s in favor of epistemic rationality imply this sort of thing.
There is more I could say about how values such as the value of staying alive can be compatible with deontological morality (of the sort compatible with will-to-think), perhaps this thread can explain some of it.
VNM/Bayes suggest there are some free parameters in how reflectively stable AGI could turn out, e.g. beliefs about completely un-testable propositions (mathematically undecidable etc), which might hypothetically be action-relevant at some point.
None of these are going to look like human values, human values aren’t reflectively stable so are distinct in quite a lot of ways. FAI is a hypothetical of a reflectively stable AGI that is nonetheless “close to” or “extended from” human values to the degree that’s possible. But it will still have very different preferences.
It would be very hard for will-to-think to be in itself “misguided”, it’s the drive to understand more, it may be compatible with other drives but without will-to-think there is no coherent epistemology or values.
Uploading is a possible path towards reflective stability that lots of people would consider aligned because it starts with a copy of them. But it’s going to look very different after millions of years of the upload’s reflection, of course. It’s going to be hard to evaluate this sort of thing on a value level because it has to be done from a perspective that doesn’t know very much, lacks reflective stability, etc.
Efficient heat engines locally slow entropy increase. If they could reverse entropy, they would (to get more energy out of things). They can also export high entropy (e.g. medium-temperature water) while intaking low entropy (e.g. un-mixed high and low temperature water) to locally reduce entropy. Entropy is waste from the perspective of a heat engine. Likewise, animals intake low-entropy food and excrete high-entropy waste.
MIRI research topics are philosophical problems. Such as decision theory and logical uncertainty. And they would have to solve more. Ontology identification is a philosophical problem. Really, how would you imagine doing FAI without solving much of philosophy?
I think the post is pretty clear about why I think it failed. MIRI axed the agent foundations team and I can see very very few people continue to work on these problems. Maybe in multiple decades (past many of the relevant people’s median superintelligence timelines) some of the problems will get solved but I don’t see “push harder on doing agent foundations” as a thing people are trying to do.
There might be a confusion. Did you get the impression from my post that I think MIRI was trying to solve philosophy?
I do think other MIRI researchers and I would think of the MIRI problems as philosophical in nature even if they’re different from the usual ones, because they’re more relevant and worth paying attention to, given the mission and so on, and because (MIRI believes) they carve philosophical reality at the joints better than the conventional ones.
Whether it’s “for the sake of solving philosophical problems or not”… clearly they think they would need to solve a lot of them to do FAI.
EDIT: for more on MIRI philosophy, see deconfusion, free will solution.
It appears Eliezer thinks executable philosophy addresses most philosophical issues worth pursuing:
Most “philosophical issues” worth pursuing can and should be rephrased as subquestions of some primary question about how to design an Artificial Intelligence, even as a matter of philosophy qua philosophy.
“Solving philosophy” is a grander marketing slogan that I don’t think was used, but, clearly, executable philosophy is a philosophically ambitious project.
None of what you’re talking about is particular to the Sequences. It’s a particular synthesis of ideas including reductionism, Bayesianism, VNM, etc. I’m not really sure why the Sequences would be important under your view except as a popularization of pre-existing concepts.
Thanks, hadn’t realized how this related to algebraic geometry. Reminds me of semi-simplicial type theory.