All of Darklight's Comments + Replies

It seems like it would depend pretty strongly on which side you view as having a closer alignment with human values generally. That probably depends a lot on your worldview and it would be very hard to be unbiased about this.

There was actually a post about almost this exact question on the EA Forums a while back. You may want to peruse some of the comments there.

Back in October 2024, I tried to test various LLM Chatbots with the question:

"Is there a way to convert a correlation to a probability while preserving the relationship 0 = 1/n?"

Years ago, I came up with an unpublished formula that does just that:

p(r) = (n^r * (r + 1)) / (2^r * n)

So I was curious if they could figure it out. Alas, back in October 2024, they all made up formulas that didn't work.

Yesterday, I tried the same question on ChatGPT and, while it didn't get it quite right, it came, very, very close. So, I modified the question to be more specific:... (read more)

The most I've seen people say "whataboutism" has been in response to someone trying to deflect criticism by pointing out apparent hypocrisy, as in the aforementioned Soviet example (I used to argue with terminally online tankies a lot). 

I.e. 

(A): "The treatment of Uyghurs in China is appalling. You should condemn this."

(B): "What about the U.S. treatment of Native Americans? Who are you to criticize?"

(A): "That's whataboutism!"

The thing I find problematic with this "defence" is that both instances are ostensibly examples of clear wrongdoing, and ... (read more)

Ok fair. I was assuming real world conditions rather than the ideal of Dath Ilan. Sorry for the confusion.

Why not? Like, the S&P 500 can vary by tens of percent, but as Google suggests, global GDP only fell 3% in 2021, and it usually grows, and the more stocks are distributed, the more stable they are.

Increases in the value of the S&P 500 are basically deflation relative to other units of account. When an asset appreciates in value, when its price goes up it is deflating relative to the currency the price is in. Like, when the price of bread increases, that means dollars are inflating, and bread is deflating. Remember, your currency is based on a perce... (read more)

I guess I'm just not sure you could trade in "hundred-trillionths of global market cap". Like, fractions of a thing assume there is still an underlying quantity or unit of measure that the fraction is a subcomponent of. If you were to range it from 0 to 1, you'd still need a way to convert a 0.0001% into a quantity of something, whether it's gold or grain or share certificates or whatever.

I can sorta imagine a fractional shares of global market cap currency coming into existence alongside other currencies that it can be exchanged for, but having all tradit... (read more)

The idea of labour hours as a unit of account isn't that new. Labour vouchers were actually tried by some utopian anarchists in the 1800s and early experiments like the Cincinnati Time Store were modestly successful. The basic idea is not to track subjective exchange values but instead a more objective kind of value, the value of labour, or a person's time, with the basic assumption that each person's time should be equally valuable. Basically, it goes back to Smith and Ricardo and the Labour Theory of Value that was popular in classical economics before m... (read more)

This put into well-written words a lot of thoughts I've had in the past but never been able to properly articulate. Thank you for writing this.

This sounds rather like the competing political economic theories of classical liberalism and Marxism to me. Both of these intellectual traditions carry a lot of complicated baggage that can be hard to disentangle from the underlying principles, but you seem to have a done a pretty good job of distilling the relevant ideas in a relatively apolitical manner.

That being said, I don't think it's necessary for these two explanations for wealth inequality to be mutually exclusive. Some wealth could be accumulated through "the means of production" as you call it,... (read more)

Another thought I just had was, could it be that ChatGPT, because it's trained to be such a people pleaser, is losing intentionally to make the user happy?

Have you tried telling it to actually try to win? Probably won't make a difference, but it seems like a really easy thing to rule out.

Also, quickly looking into how LLM token sampling works nowadays, you may also need to set the parameters top_p to 0, and top_k to 1 to get it to actually function like argmax. Looks like these can only be set through the API if you're using ChatGPT or similar proprietary LLMs. Maybe I'll try experimenting with this when I find the time, if nothing else to rule out the possibility of such a seemingly obvious thing being missed.

I've always wondered with these kinds of weird apparent trivial flaws in LLM behaviour if it doesn't have something to do with the way the next token is usually randomly sampled from the softmax multinomial distribution rather than taking the argmax (most likely) of the probabilities. Does anyone know if reducing the temperature parameter to zero so that it's effectively the argmax changes things like this at all?

p = (n^c * (c + 1)) / (2^c * n)

As far as I know, this is unpublished in the literature. It's a pretty obscure use case, so that's not surprising. I have doubts I'll ever get around to publishing the paper I wanted to write that uses this in an activation function to replace softmax in neural nets, so it probably doesn't matter much if I show it here.

So, my main idea is that the principle of maximum entropy aka the principle of indifference suggests a prior of 1/n where n is the number of possibilities or classes. P x 2 - 1 leads to p = 0.5 for c = 0. What I want is for c = 0 to lead to p = 1/n rather than 0.5, so that it works in the multiclass cases where n is greater than 2.

Correlation space is between -1 and 1, with 1 being the same (definitely true), -1 being the opposite (definitely false), and 0 being orthogonal (very uncertain). I had the idea that you could assume maximum uncertainty to be 0 in correlation space, and 1/n (the uniform distribution) in probability space.

I tried asking ChatGPT, Gemini, and Claude to come up with a formula that converts between correlation space to probability space while preserving the relationship 0 = 1/n. I came up with such a formula a while back, so I figure it shouldn't be hard. They all offered formulas, all of which were shown to be very much wrong when I actually graphed them to check.

I was not aware of these. Thanks!

Thanks for the clarifications. My naive estimate is obviously just a simplistic ballpark figure using some rough approximations, so I appreciate adding some precision.

Also, even if we can train and run a model the size of the human brain, it would still be many orders of magnitude less energy efficient than an actual brain. Human brains use barely 20 watts. This hypothetical GPU brain would require enormous data centres of power, and each H100 GPU uses 700 watts alone.

I've been looking at the numbers with regards to how many GPUs it would take to train a model with as many parameters as the human brain has synapses. The human brain has 100 trillion synapses, and they are sparse and very efficiently connected. A regular AI model fully connects every neuron in a given layer to every neuron in the previous layer, so that would be less efficient.

The average H100 has 80 GB of VRAM, so assuming that each parameter is 32 bits, then you have about 20 billion per GPU. So, you'd need 10,000 GPUs to fit a single instance of a huma... (read more)

I ran out of the usage limit for GPT-4o (seems to just be 10 prompts every 5 hours) and it switched to GPT-4o-mini. I tried asking it the Alpha Omega question and it made some math nonsense up, so it seems like the model matters for this for some reason.

So, a while back I came up with an obscure idea I called the Alpha Omega Theorem and posted it on the Less Wrong forums. Given how there's only one post about it, it shouldn't be something that LLMs would know about. So in the past, I'd ask them "What is the Alpha Omega Theorem?", and they'd always make up some nonsense about a mathematical theory that doesn't actually exist. More recently, Google Gemini and Microsoft Bing Chat would use search to find my post and use that as the basis for their explanation. However, I only have the free version of ChatGPT... (read more)

I'm wondering what people's opinions are on how urgent alignment work is. I'm a former ML scientist who previously worked at Maluuba and Huawei Canada, but switched industries into game development, at least in part to avoid contributing to AI capabilities research. I tried earlier to interview with FAR and Generally Intelligent, but didn't get in. I've also done some cursory independent AI safety research in interpretability and game theoretic ideas my spare time, though nothing interesting enough to publish yet.

My wife also recently had a baby, and carin... (read more)

Thanks for the reply!

So, the main issue I'm finding with putting them all into one proposal is that there's a 1000 character limit on the main summary section where you describe the project, and I cannot figure out how to cram multiple ideas into that 1000 characters without seriously compromising the quality of my explanations for each.

I'm not sure if exceeding that character limit will get my proposal thrown out without being looked at though, so I hesitate to try that. Any thoughts?

I already tried discussing a very similar concept I call Superrational Signalling in this post. It got almost no attention, and I have doubts that Less Wrong is receptive to such ideas.

I also tried actually programming a Game Theoretic simulation to try to test the idea, which you can find here, along with code and explanation. Haven't gotten around to making a full post about it though (just a shortform).

So, I have three very distinct ideas for projects that I'm thinking about applying to the Long Term Future Fund for. Does anyone happen to know if it's better to try to fit them all into one application, or split them into three separate applications?

Recently I tried out an experiment using the code from the Geometry of Truth paper to try to see if using simple label words like "true" and "false" could substitute for the datasets used to create truth probes. I also tried out a truth probe algorithm based on classifying with the higher cosine similarity to the mean vectors.

Initial results seemed to suggest that the label word vectors were sorta acceptable, albeit not nearly as good (around 70% accurate rather than 95%+ like with the datasets). However, testing on harder test sets showed much worse accur... (read more)

Update: I made an interactive webpage where you can run the simulation and experiment with a different payoff matrix and changes to various other parameters.

So, I adjusted the aggressor system to work like alliances or defensive pacts instead of a universal memory tag. Basically, now players make allies when they both cooperate and aren't already enemies, and make enemies when defected against first, which sets all their allies to also consider the defector an enemy. This, doesn't change the result much. The alliance of nice strategies still wins the vast majority of the time.

I also tried out false flag scenarios where 50% of the time the victim of a defect first against non-enemy will actually be mistaken for... (read more)

Admittedly this is a fairly simple set up without things like uncertainty and mistakes, so yes, it may not really apply to the real world. I just find it interesting that it implies that strong coordinated retribution can, at least in this toy set up, be useful for shaping the environment into one where cooperation thrives, even after accounting for power differentials and the ability to kill opponents outright, which otherwise change the game enough that straight Tit-For-Tat doesn't automatically dominate.

It's possible there are some situations where this... (read more)

Okay, so I decided to do an experiment in Python code where I modify the Iterated Prisoner's Dilemma to include Death, Asymmetric Power, and Aggressor Reputation, and run simulations to test how different strategies do. Basically, each player can now die if their points falls to zero or below, and the payoff matrix uses their points as a variable such that there is a power difference that affects what happens. Also, if a player defects first in any round of any match against a non-aggressor, they get the aggressor label, which matters for some strategies t... (read more)

I was recently trying to figure out a way to calculate my P(Doom) using math. I initially tried just making a back of the envelope calculation by making a list of For and Against arguments and then dividing the number of For arguments by the total number of arguments. This led to a P(Doom) of 55%, which later got revised to 40% when I added more Against arguments. I also looked into using Bayes Theorem and actual probability calculations, but determining P(E | H) and P(E) to input into P(H | E) = P(E | H) * P(H) / P(E) is surprisingly hard and confusing.

Minor point, but the apology needs to sound sincere and credible, usually by being specific about the mistakes and concise and to the point and not like, say, Bostrom's defensive apology about the racist email a while back. Otherwise you can instead signal that you are trying to invoke the social API call in a disingenuous way, which can clearly backfire.

Things like "sorry you feel offended" also tend to sound like you're not actually remorseful for your actions and are just trying to elicit the benefits of an apology. None of the apologies you described sound anything like that, but it's a common failure state among the less emotionally mature and the syncophantic.

I have some ideas and drafts for posts that I've been sitting on because I feel somewhat intimidated by the level of intellectual rigor I would need to put into the final drafts to ensure I'm not downvoted into oblivion (something a younger me experienced in the early days of Less Wrong).

Should I try to overcome this fear, or is it justified?

For instance, I have a draft of a response to Eliezer's List of Lethalities post that I've been sitting on since 2022/04/11 because I doubted it would be well received given that it tries to be hopeful and, as a former... (read more)

I still remember when I was a masters student presenting a paper at the Canadian Conference on AI 2014 in Montreal and Bengio was also at the conference presenting a tutorial, and during the Q&A afterwards, I asked him a question about AI existential risk. I think I worded it back then as concerned about the possibility of Unfriendly AI or a dangerous optimization algorithm or something like that, as it was after I'd read the sequences but before "existential risk" was popularized as a term. Anyway, he responded by asking jokingly if I was a journalist... (read more)

I recently interviewed with Epoch, and as part of a paid work trial they wanted me to write up a blog post about something interesting related to machine learning trends. This is what I came up with:

http://www.josephius.com/2022/09/05/energy-efficiency-trends-in-computation-and-long-term-implications/

I should point out that the logic of the degrowth movement follows from a relatively straightforward analysis of available resources vs. first world consumption levels.  Our world can only sustain 7 billion human beings because the vast majority of them live not at first world levels of consumption, but third world levels, which many would argue to be unfair and an unsustainable pyramid scheme.  If you work out the numbers, if everyone had the quality of life of a typical American citizen, taking into account things like meat consumption to arabl... (read more)

I'm using the number calculated by Ray Kurzweil for his book, the Age of Spiritual Machines from 1999.  To get that figure, you need 100 billion neurons firing every 5 ms, or 200 Hz.  That is based on the maximum firing rate given refractory periods.  In actuality, average firing rates are usually lower than that, so in all likelihood the difference isn't actually six orders of magnitude.  In particular, I should point out that six orders of magnitude is referring to the difference between this hypothetical maximum firing brain and the ... (read more)

Okay, so I contacted 80,000 hours, as well as some EA friends for advice.  Still waiting for their replies.

I did hear from an EA who suggested that if I don't work on it, someone else who is less EA-aligned will take the position instead, so in fact, it's slightly net positive for myself to be in the industry, although I'm uncertain whether or not AI capability is actually funding constrained rather than personal constrained.

Also, would it be possible to mitigate the net negative by choosing to deliberately avoid capability research and just take an ML engineering job at a lower tier company that is unlikely to develop AGI before others and just work on applying existing ML tech to solving practical problems?

I previously worked as a machine learning scientist but left the industry a couple of years ago to explore other career opportunities.  I'm wondering at this point whether or not to consider switching back into the field.  In particular, in case I cannot find work related to AI safety, would working on something related to AI capability be a net positive or net negative impact overall?

Even further research shows the most recent Nvidia RTX 3090 is actually slightly more efficient than the 1660 Ti, at 36 TeraFlops, 350 watts, and 2.2 kg, which works out to 0.0001 PetaFlops/Watt and 0.016 PetaFlops/kg.  Once again, they're within an order of magnitude of the supercomputers.

So, I did some more research, and the general view is that GPUs are more power efficient in terms of Flops/watt than CPUs, and the most power efficient of those right now is the Nvidia 1660 Ti, which comes to 11 TeraFlops at 120 watts, so 0.000092 PetaFlops/Watt, which is about 6x more efficient than Fugaku.  It also weighs about 0.87 kg, which works out to 0.0126 PetaFlops/kg, which is about 7x more efficient than Fugaku.  These numbers are still within an order of magnitude, and also don't take into account the overhead costs of things like coo... (read more)

Another thought is that maybe Less Wrong itself, if it were to expand in size and become large enough to roughly represent humanity, could be used as such a dataset.

So, I had a thought.  The glory system idea that I posted about earlier, if it leads to a successful, vibrant democratic community forum, could actually serve as a kind of dataset for value learning.  If each post has a number attached to it that indicates the aggregated approval of human beings, this can serve as a rough proxy for a kind of utility or Coherent Aggregated Volition.

Given that individual examples will probably be quite noisy, but averaged across a large amount of posts, it could function as a real world dataset, with the post conte... (read more)

A further thought is that those with more glory can be seen almost as elected experts.  Their glory is assigned to them by votes after all.  This is an important distinction from an oligarchy.  I would actually be inclined to see the glory system as located on a continuum between direct demcracy and representative democracy.

So, keep in mind that by having the first vote free and worth double the paid votes does tilt things more towards democracy.  That being said, I am inclined to see glory as a kind of proxy for past agreement and merit, and a rough way to approximate liquid democracy where you can proxy your vote to others or vote yourself.

In this alternative "market of ideas" the ideas win out because people who others trust to have good opinions are able to leverage that trust.  Decisions over the merit of the given arguments are aggregated by vote.  As lon... (read more)

Perhaps a nitpick detail, but having someone rob them would not be equivalent, because the cost of the action is offset by the ill-gotten gains.  The proposed currency is more directly equivalent to paying someone to break into the target's bank account and destroying their assets by a proportional amount so that no one can use them anymore.

As for the more general concerns:

Standardized laws and rules tend in practice to disproportionately benefit those with the resources to bend and manipulate those rules with lawyers.  Furthermore, this proposal... (read more)

As for the cheaply punishing prolific posters problem, I don't know a good solution that doesn't lead to other problems, as forcing all downvotes to cost glory makes it much harder to deal with spammers who somehow get through the application process filter.  I had considered an alternative system in which all votes cost glory, but then there's no way to generate glory except perhaps by having admins and mods gift them, which could work, but runs counter to the direct democracy ideal that I was sorta going for.