mind uploading

All posts tagged mind uploading

I’ve been dedicating a fair amount of my time recently to investigating whole brain emulation (WBE).

As computational power continues to grow, the feasibility of emulating a human brain at a reasonable speed becomes increasingly plausible.

While the connectome data alone seems insufficient to fully capture and replicate human behavior, recent advancements in scanning technology have provided valuable insights into distinguishing different types of neural connections. I’ve heard suggestions that combining this neuron-scale data with higher-level information, such as fMRI or EEG, might hold the key to unlocking WBE. However, the evidence is not yet conclusive enough for me to make any definitive statements.

I’ve heard some talk about a new company aiming to achieve WBE within the next five years. While this timeline aligns suspiciously with the typical venture capital horizon for industries with weak patent protection, I believe there is a non-negligible chance of success within the next decade – perhaps exceeding 10%. As a result, I’m actively exploring investment opportunities in this company.

There has also been speculation about the potential of WBE to aid in AI alignment efforts. However, I remain skeptical about this prospect. For WBE to make a significant impact on AI alignment, it would require not only an acceleration in WBE progress but also a slowdown in AI capability advances as they approach human levels or the assumption that the primary risks from AI emerge only when it substantially surpasses human intelligence.

My primary motivation for delving into WBE stems from a personal desire to upload my own mind. The potential benefits of WBE for those who choose not to upload remain uncertain, and I’m uncertain how to predict its broader societal implications.

Here are some videos that influenced my recent increased interest. Note that I’m relying heavily on the reputations of the speakers when deciding how much weight to give to their opinions.

Some relevant prediction markets:

Additionally, I’ve been working on some of the suggestions mentioned in the first video. I’m sharing my code and analysis on Colab. My aim is to evaluate the resilience of language models to the types of errors that might occur during the brain scanning process. While the results provide some reassurance, their value heavily relies on assumptions about the importance of low-confidence guesses made by the emulated mind.

Or, why I don’t fear the p-zombie apocalypse.

This post analyzes concerns about how evolution, in the absence of a powerful singleton, might, in the distant future, produce what Nick Bostrom calls a “Disneyland without children”. I.e. a future with many agents, whose existence we don’t value because they are missing some important human-like quality.

The most serious description of this concern is in Bostrom’s The Future of Human Evolution. Bostrom is cautious enough that it’s hard to disagree with anything he says.

Age of Em has prompted a batch of similar concerns. Scott Alexander at SlateStarCodex has one of the better discussions (see section IV of his review of Age of Em).

People sometimes sound like they want to use this worry as an excuse to oppose the age of em scenario, but it applies to just about any scenario with human-in-a-broad-sense actors. If uploading never happens, biological evolution could produce slower paths to the same problem(s) [1]. Even in the case of a singleton AI, the singleton will need to solve the tension between evolution and our desire to preserve our values, although in that scenario it’s more important to focus on how the singleton is designed.

These concerns often assume something like the age of em lasts forever. The scenario which Age of Em analyzes seems unstable, in that it’s likely to be altered by stranger-than-human intelligence. But concerns about evolution only depend on control being sufficiently decentralized that there’s doubt about whether a central government can strongly enforce rules. That situation seems sufficiently stable to be worth analyzing.

I’ll refer to this thing we care about as X (qualia? consciousness? fun?), but I expect people will disagree on what matters for quite some time. Some people will worry that X is lost in uploading, others will worry that some later optimization process will remove X from some future generation of ems.

I’ll first analyze scenarios in which X is a single feature (in the sense that it would be lost in a single step). Later, I’ll try to analyze the other extreme, where X is something that could be lost in millions of tiny steps. Neither extreme seems likely, but I expect that analyzing the extremes will illustrate the important principles.

Continue Reading

Will young ems be created? Why and how will it happen?

Any children that exist as ems will be important as em societies mature, because they will adapt better to em environments than ems who uploaded as adults, making them more productive.

The Age of Em says little about children, presumably in part because no clear outside view predictions seem possible.

This post will use a non-Hansonian analysis style to speculate about which children will become ems. I’m writing this post to clarify my more speculative thoughts about how em worlds will work, without expecting to find much evidence to distinguish the good ideas from the bad ones.

Robin predicts few regulatory obstacles to uploading children, because he expects the world to be dominated by ems. I’m skeptical of that. Ems will be dominant in the sense of having most of the population, but that doesn’t tell us much about em influence on human society – farmers became a large fraction of the world population without meddling much in hunter-gatherer political systems. And it’s unclear whether em political systems would want to alter the relevant regulations – em societies will have much the same conflicting interest groups pushing for and against immigration that human societies have.

How much of Robin’s prediction of low regulation is due to his desire to start by analyzing a relatively simple scenario (low regulation) and add complexity later?

Continue Reading

One of most important assumptions in The Age of Ems is that non-em AGI will take a long time to develop.

1.

Scott Alexander at SlateStarCodex complains that Robin rejects survey data that uses validated techniques, and instead uses informal surveys whose results better fit Robin’s biases [1]. Robin clearly explains one reason why he does that: to get the outside view of experts.

Whose approach to avoiding bias is better?

  • Minimizing sampling error and carefully documenting one’s sampling technique are two of the most widely used criteria to distinguish science from wishful thinking.
  • Errors due to ignoring the outside view have been documented to be large, yet forecasters are reluctant to use the outside view.

So I rechecked advice from forecasting experts such as Philip Tetlock and Nate Silver, and the clear answer I got was … that was the wrong question.

Tetlock and Silver mostly focus on attitudes that are better captured by the advice to be a fox, not a hedgehog.

The strongest predictor of rising into the ranks of superforecasters is perpetual beta, the degree to which one is committed to belief updating and self-improvement.

Tetlock’s commandment number 3 says “Strike the right balance between inside and outside views”. Neither Tetlock or Silver offer hope that either more rigorous sampling of experts or dogmatically choosing the outside view over the inside view help us win a forecasting contest.

So instead of asking who is right, we should be glad to have two approaches to ponder, and should want more. (Robin only uses one approach for quantifying the time to non-em AGI, but is more fox-like when giving qualitative arguments against fast AGI progress).

2.

What Robin downplays is that there’s no consensus of the experts on whom he relies, not even about whether progress is steady, accelerating, or decelerating.

Robin uses the median expert estimate of progress in various AI subfields. This makes sense if AI progress depends on success in many subfields. It makes less sense if success in one subfield can make the other subfields obsolete. If “subfield” means a guess about what strategy best leads to intelligence, then I expect the median subfield to be rendered obsolete by a small number of good subfields [2]. If “subfield” refers to a subset of tasks that AI needs to solve (e.g. vision, or natural language processing), then it seems reasonable to look at the median (and I can imagine that slower subfields matter more). Robin appears to use both meanings of “subfield”, with fairly similar results for each, so it’s somewhat plausible that the median is informative.

3.

Scott also complains that Robin downplays the importance of research spending while citing only a paper dealing with government funding of agricultural research. But Robin also cites another paper (Ulku 2004), which covers total R&D expenditures in 30 countries (versus 16 countries in the paper that Scott cites) [3].

4.

Robin claims that AI progress will slow (relative to economic growth) due to slowing hardware progress and reduced dependence on innovation. Even if I accept Robin’s claims about these factors, I have trouble believing that AI progress will slow.

I expect higher em IQ will be one factor that speeds up AI progress. Garrett Jones suggests that a 40 IQ point increase in intelligence causes a 50% increase in a country’s productivity. I presume that AI researcher productivity is more sensitive to IQ than is, say, truck driver productivity. So it seems fairly plausible to imagine that increased em IQ will cause more than a factor of two increase in the rate of AI progress. (Robin downplays the effects of IQ in contexts where a factor of two wouldn’t much affect his analysis; he appears to ignore them in this context).

I expect that other advantages of ems will contribute additional speedups – maybe ems who work on AI will run relatively fast, maybe good training/testing data will be relatively cheap to create, or maybe knowledge from experimenting on ems will better guide AI research.

5.

Robin’s arguments against an intelligence explosion are weaker than they appear. I mostly agree with those arguments, but I want to discourage people from having strong confidence in them.

The most suspicious of those arguments is that gains in software algorithmic efficiency “remain surprisingly close to the rate at which hardware costs have fallen. This suggests that algorithmic gains have been enabled by hardware gains”. He cites only (Grace 2013) in support of this. That paper doesn’t comment on whether hardware changes enable software changes. The evidence seems equally consistent with that or with the hypothesis that both are independently caused by some underlying factor. I’d say there’s less than a 50% chance that Robin is correct about this claim.

Robin lists 14 other reasons for doubting there will be an intelligence explosion: two claims about AI history (no citations), eight claims about human intelligence (one citation), and four about what causes progress in research (with the two citations mentioned earlier). Most of those 14 claims are probably true, but it’s tricky to evaluate their relevance.

Conclusion

I’d say there’s maybe a 15% chance that Robin is basically right about the timing of non-em AI given his assumptions about ems. His book is still pretty valuable if an em-dominated world lasts for even one subjective decade before something stranger happens. And “something stranger happens” doesn’t necessarily mean his analysis becomes obsolete.

Footnotes

[1] – I can’t find any SlateStarCodex complaint about Bostrom doing something in Superintelligence that’s similar to what Scott accuses Robin of, when Bostrom’s survey of experts shows an expected time of decades for human-level AI to become superintelligent. Bostrom wants to focus on a much faster takeoff scenario, and disagrees with the experts, without identifying reasons for thinking his approach reduces biases.

[2] – One example is that genetic algorithms are looking fairly obsolete compared to neural nets, now that they’re being compared on bigger problems than when genetic algorithms were trendy.

Robin wants to avoid biases from recent AI fads by looking at subfields as they were defined 20 years ago. Some recent changes in AI are fads, but some are increased wisdom. I expect many subfields to be dead ends, given how immature AI was 20 years ago (and may still be today).

[3] – Scott quotes from one of three places that Robin mentions this subject (an example of redundancy that is quite rare in the book), and that’s the one place out of three where Robin neglects to cite (Ulku 2004). Age of Em is the kind of book where it’s easy to overlook something important like that if you don’t read it more carefully than you’d read a normal book.

I tried comparing (Ulku 2004) to the OECD paper that Scott cites, and failed to figure out whether they disagree. The OECD paper is probably consistent with Robin’s “less than proportionate increases” claim that Scott quotes. But Scott’s doubts are partly about Robin’s bolder prediction that AI progress will slow down, and academic papers don’t help much in evaluating that prediction.

If you’re tempted to evaluate how well the Ulku paper supports Robin’s views, beware that this quote is one of its easier to understand parts:

In addition, while our analysis lends support for endogenous growth theories in that it confirms a significant relationship between R&D stock and innovation, and between innovation and per capita GDP, it lacks the evidence for constant returns to innovation in terms of R&D stock. This implies that R&D models are not able to explain sustainable economic growth, i.e. they are not fully endogenous.

Book review: The Age of Em: Work, Love and Life when Robots Rule the Earth, by Robin Hanson.

This book analyzes a possible future era when software emulations of humans (ems) dominate the world economy. It is too conservative to tackle longer-term prospects for eras when more unusual intelligent beings may dominate the world.

Hanson repeatedly tackles questions that scare away mainstream academics, and gives relatively ordinary answers (guided as much as possible by relatively standard, but often obscure, parts of the academic literature).

Assumptions

Hanson’s scenario relies on a few moderately controversial assumptions. The assumptions which I find most uncertain are related to human-level intelligence being hard to understand (because it requires complex systems), enough so that ems will experience many subjective centuries before artificial intelligence is built from scratch. For similar reasons, ems are opaque enough that it will be quite a while before they can be re-engineered to be dramatically different.

Hanson is willing to allow that ems can be tweaked somewhat quickly to produce moderate enhancements (at most doubling IQ) before reaching diminishing returns. He gives somewhat plausible reasons for believing this will only have small effects on his analysis. But few skeptics will be convinced.

Some will focus on potential trillions of dollars worth of benefits that higher IQs might produce, but that wealth would not much change Hanson’s analysis.

Others will prefer an inside view analysis which focuses on the chance that higher IQs will better enable us to handle risks of superintelligent software. Hanson’s analysis implies we should treat that as an unlikely scenario, but doesn’t say what we should do about modest probabilities of huge risks.

Another way that Hanson’s assumptions could be partly wrong is if tweaking the intelligence of emulated Bonobos produces super-human entities. That seems to only require small changes to his assumptions about how tweakable human-like brains are. But such a scenario is likely harder to analyze than Hanson’s scenario, and it probably makes more sense to understand Hanson’s scenario first.

Wealth

Wages in this scenario are somewhat close to subsistence levels. Ems have some ability to restrain wage competition, but less than they want. Does that mean wages are 50% above subsistence levels, or 1%? Hanson hints at the former. The difference feels important to me. I’m concerned that sound-bite versions of book will obscure the difference.

Hanson claims that “wealth per em will fall greatly”. It would be possible to construct a measure by which ems are less wealthy than humans are today. But I expect it will be at least as plausible to use a measure under which ems are rich compared to humans of today, but have high living expenses. I don’t believe there’s any objective unit of value that will falsify one of those perspectives [1].

Style / Organization

The style is more like a reference book than a story or an attempt to persuade us of one big conclusion. Most chapters (except for a few at the start and end) can be read in any order. If the section on physics causes you to doubt whether the book matters, skip to chapter 12 (labor), and return to the physics section later.

The style is very concise. Hanson rarely repeats a point, so understanding him requires more careful attention than with most authors.

It’s odd that the future of democracy gets less than twice as much space as the future of swearing. I’d have preferred that Hanson cut out a few of his less important predictions, to make room for occasional restatements of important ideas.

Many little-known results that are mentioned in the book are relevant to the present, such as: how the pitch of our voice affects how people perceive us, how vacations affect productivity, and how bacteria can affect fluid viscosity.

I was often tempted to say that Hanson sounds overconfident, but he is clearly better than most authors at admitting appropriate degrees of uncertainty. If he devoted much more space to caveats, I’d probably get annoyed at the repetition. So it’s hard to say whether he could have done any better.

Conclusion

Even if we should expect a much less than 50% chance of Hanson’s scenario becoming real, it seems quite valuable to think about how comfortable we should be with it and how we could improve on it.

Footnote

[1] – The difference matters only in one paragraph, where Hanson discusses whether ems deserve charity more than do humans living today. Hanson sounds like he’s claiming ems deserve our charity because they’re poor. Most ems in this scenario are comfortable enough for this to seem wrong.

Hanson might also be hinting that our charity would be effective at increasing the number of happy ems, and that basic utilitarianism says that’s preferable to what we can do by donating to today’s poor. That argument deserves more respect and more detailed analysis.

Connectomes are not sufficient by themselves to model brain behavior. Brain modeling has been limited more by the need for good information about the dynamic behavior of individual neurons.

The paper Whole-brain calcium imaging with cellular resolution in freely behaving Caenorhabditis elegans looks like an important step toward overcoming this limitation. The authors observed the behavior of many individual neurons in a moving nematode.

They still can’t reliably map the neurons they observed to standard C. elegans neuron names:

The neural position validation experiments presented here, however, have led us to conclude that worm-to-worm variability in neuronal position in the head is large enough to pose a formidable challenge for neuron identification.

But there are enough hints about which neurons do what that I’m confident this problem can be solved if enough effort is devoted to it.

My biggest uncertainty concerns applying this approach to mammalian brains. Mammalian brains aren’t transparent enough to be imaged this way. Are C. elegans neurons similar enough that we can just apply the same models to both? I suspect not.

I’d like to see more discussion of uploaded ape risks.

There is substantial disagreement over how fast an uploaded mind (em) would improve its abilities or the abilities of its progeny. I’d like to start by analyzing a scenario where it takes between one and ten years for an uploaded bonobo to achieve human-level cognitive abilities. This scenario seems plausible, although I’ve selected it more to illustrate a risk that can be mitigated than because of arguments about how likely it is.

I claim we should anticipate at least a 20% chance a human-level bonobo-derived em would improve at least as quickly as a human that uploaded later.

Considerations that weigh in favor of this are: that bonobo minds seem to be about as general-purpose as humans, including near-human language ability; and the likely ease of ems interfacing with other software will enable them to learn new skills faster than biological minds will.

The most concrete evidence that weighs against this is the modest correlation between IQ and brain size. It’s somewhat plausible that it’s hard to usefully add many neurons to an existing mind, and that bonobo brain size represents an important cognitive constraint.

I’m not happy about analyzing what happens when another species develops more powerful cognitive abilities than humans, so I’d prefer to have some humans upload before the bonobos become superhuman.

A few people worry that uploading a mouse brain will generate enough understanding of intelligence to quickly produce human-level AGI. I doubt that biological intelligence is simple / intelligible enough for that to work. So I focus more on small tweaks: the kind of social pressures which caused the Flynn Effect in humans, selective breeding (in the sense of making many copies of the smartest ems, with small changes to some copies), and faster software/hardware.

The risks seem dependent on the environment in which the ems live and on the incentives that might drive their owners to improve em abilities. The most obvious motives for uploading bonobos (research into problems affecting humans, and into human uploading) create only weak incentives to improve the ems. But there are many other possibilities: military use, interesting NPCs, or financial companies looking for interesting patterns in large databases. No single one of those looks especially likely, but with many ways for things to go wrong, the risks add up.

What could cause a long window between bonobo uploading and human uploading? Ethical and legal barriers to human uploading, motivated by risks to the humans being uploaded and by concerns about human ems driving human wages down.

What could we do about this risk?

Political activism may mitigate the risks of hostility to human uploading, but if done carelessly it could create a backlash which worsens the problem.

Conceivably safety regulations could restrict em ownership/use to people with little incentive to improve the ems, but rules that looked promising would still leave me worried about risks such as irresponsible people hacking into computers that run ems and stealing copies.

A more sophisticated approach is to improve the incentives to upload humans. I expect the timing of the first human uploads to be fairly sensitive to whether we have legal rules which enable us to predict who will own em labor. But just writing clear rules isn’t enough – how can we ensure political support for them at a time when we should expect disputes over whether they’re people?

We could also find ways to delay ape uploading. But most ways of doing that would also delay human uploading, which creates tradeoffs that I’m not too happy with (partly due to my desire to upload before aging damages me too much).

If a delay between bonobo and human uploading is dangerous, then we should also ask about dangers from other uploaded species. My intuition says the risks are much lower, since it seems like there are few technical obstacles to uploading a bonobo brain shortly after uploading mice or other small vertebrates.

But I get the impression that many people associated with MIRI worry about risks of uploaded mice, and I don’t have strong evidence that I’m wiser than they are. I encourage people to develop better analyses of this issue.

Ken Hayworth has created an interesting prize for Brain Preservation Technology, designed to improve techniques of relevance to cryonics and mind uploading, but intended to be relevant to goals that don’t require preserving individual identity (such as better understanding of generic brains).

Many of the prize criteria are well thought out, especially the ones concerning quality of preservation. But there a few criteria for which it’s hard to predict how the judges would evaluate a proposed technique, and they will significantly impair the effectiveness of the prize.

The requirement that it have the potential to be performed for less than $20,000 requires a number of subjective judgments, such as the cost of training the necessary personnel (which will be affected by the quality of the trainers and trainees).

The requirement that it “be absolutely safe for the personnel involved” would seem to be prohibitive if I try to interpret it literally. A somewhat clearer approach would be to require that it be at least as safe as some commonly preformed procedure. But the effort required to compare risks will be far from trivial.

The requirement that we have reason to expect the preserved brains to remain stable for 100 years depends on some assumptions that aren’t well explained, such as why a shorter time period wouldn’t be enough (which depends on the specific goals of preservation and on predictions about how fast technology progresses), and what we should look at to estimate the durability – I suspect the obstacles to long-term stability are different for different techniques.

(I noticed this prize in connection with the ASIM 2010 conference, although I didn’t get much out of the part of the conference that I was able to attend).

Some comments on last weekend’s Foresight Conference:

At lunch on Sunday I was in a group dominated by a discussion between Robin Hanson and Eliezer Yudkowsky over the relative plausibility of new intelligences having a variety of different goal systems versus a single goal system (as in a society of uploads versus Friendly AI). Some of the debate focused on how unified existing minds are, with Eliezer claiming that dogs mostly don’t have conflicting desires in different parts of their minds, and Robin and others claiming such conflicts are common (e.g. when deciding whether to eat food the dog has been told not to eat).

One test Eliezer suggested for the power of systems with a unified goal system is that if Robin were right, bacteria would have outcompeted humans. That got me wondering whether there’s an appropriate criterion by which humans can be said to have outcompeted bacteria. The most obvious criterion on which humans and bacteria are trying to compete is how many copies of their DNA exist. Using biomass as a proxy, bacteria are winning by several orders of magnitude. Another possible criterion is impact on large-scale features of Earth. Humans have not yet done anything that seems as big as the catastrophic changes to the atmosphere (“the oxygen crisis”) produced by bacteria. Am I overlooking other appropriate criteria?

Kartik Gada described two humanitarian innovation prizes that bear some resemblance to a valuable approach to helping the world’s poorest billion people, but will be hard to turn into something with a reasonable chance of success. The Water Liberation Prize would be pretty hard to judge. Suppose I submit a water filter that I claim qualifies for the prize. How will the judges test the drinkability of the water and the reusability of the filter under common third world conditions (which I suspect vary a lot and which probably won’t be adequately duplicated where the judges live)? Will they ship sample devices to a number of third world locations and ask whether it produces water that tastes good, or will they do rigorous tests of water safety? With a hoped for prize of $50,000, I doubt they can afford very good tests. The Personal Manufacturing Prizes seem somewhat more carefully thought out, but need some revision. The “three different materials” criterion is not enough to rule out overly specialized devices without some clear guidelines about which differences are important and which are trivial. Setting specific award dates appears to assume an implausible ability to predict how soon such a device will become feasible. The possibility that some parts of the device are patented is tricky to handle, as it isn’t cheap to verify the absence of crippling patents.

There was a debate on futarchy between Robin Hanson and Mencius Moldbug. Moldbug’s argument seems to boil down to the absence of a guarantee that futarchy will avoid problems related to manipulation/conflicts of interest. It’s unclear whether he thinks his preferred form of government would guarantee any solution to those problems, and he rejects empirical tests that might compare the extent of those problems under the alternative systems. Still, Moldbug concedes enough that it should be possible to incorporate most of the value of futarchy within his preferred form of government without rejecting his views. He wants to limit trading to the equivalent of the government’s stockholders. Accepting that limitation isn’t likely to impair the markets much, and may make futarchy more palatable to people who share Moldbug’s superstitions about markets.

Some of Robin Hanson’s Malthusian-sounding posts prompted me to wonder how we can create a future that is better than the repugnant conclusion. It struck me that there’s no reason to accept the assumption that increasing the number of living minds to the limit of available resources implies that the quality of the lives those minds live will decrease to where they’re barely worth living.

If we imagine the minds to be software, then a mind that barely has enough resources to live could be designed so that it is very happy with the cpu cycles or negentropy it gets even if those are negligible compared to other minds. Or if there is some need for life to be biological, a variant of hibernation might accomplish the same result.

If this is possible, then what I find repugnant about the repugnant conclusion is that it perpetuates the cruelty of evolution which produces suffering in beings with fewer resources than they were evolved to use. Any respectable civilization will engineer away the conflict between average utilitarianism and total utilitarianism.

If instead the most important limit on the number of minds is the supply of matter, then there is a tradeoff between more minds and more atoms per mind. But there is no mere addition paradox to create concerns about a repugnant conclusion if the creation of new minds reduces the utility of other minds.

(Douglas W. Portmore has a similar but less ambitious conclusion (pdf)).