JBlack

Karma: 1,836

JBlack 5 Jun 2023 0:27 UTC
36 points
32
on: I bet everyone 1000€ that I can make them dramatically happier & cure their depression in 3 months!
This doesn’t look like a bet. It looks like a service for which you charge €3500 and 3+ months of the customer’s time, but will refund €2000 of that if they don’t think you lived up to your claims.

JBlack 11 May 2022 0:17 UTC
31 points
on: Crises Don’t Need Your Software
But even if the charity you pick only yields 1 euro of value from the 1000 you send, it’s still be better to send the money.
That’s up to your own personal utility function of course, but another option is to donate somewhere else instead. There are still a great many causes that can do a lot more good for a lot more people in similarly dire situations than getting €1 worth of help out of every €1000 to the people who need it.
Yes, this sucks for people who are in terrible situations where helping them is also orders of magnitude more inefficient and/or dangerous to those trying to help than others in similar need. Maybe it feels better to spend your €1000 to give €1 of help to someone in such a situation, than to give hundreds of times more help to people in situations elsewhere that are just as bad but more accessible.

JBlack 7 Oct 2021 3:29 UTC
26 points
on: Considerations on Compensation
There seems to be a huge blind spot here, dismissed in just one sentence:
I don’t think more money moves the needle for anyone who’s familiar with a cause area but isn’t convinced it’s worth working on or directing resources to.
Of course it moves the needle! That’s how pretty much everything in the world outside direct family and friend relationships gets done.
There isn’t even a category in this article for people who would never donate, but will still do good work for pay. That covers the vast majority of the people who could be working for the organization! It’s a huge blind spot to completely ignore these people who almost certainly make up 99% or more of the talent pool!
The only way it makes sense to just ignore all those people is if the talent pool is so broad and deep that you can afford to just exclude almost all of them, or if the organization is based on some ideology that absolutely requires converts instead of just people who will do good work.
Even among people who think “this organization’s work is worthwhile”, there’s a huge gulf to “I’ll donate much of my disposable income to this specific organization” in the form of being paid below the market rate. An organization that requires the latter is definitely going to get a lot fewer candidates, even among those who think that the organization is doing worthwhile things and would donate to it.

JBlack 18 Apr 2023 0:22 UTC
23 points
18
in reply to: Raemon’s comment on: Moderation notes re: recent Said/Duncan threads
Is there any evidence that either Duncan or Said are actually detrimental to the site in general, or is it mostly in their interactions directly with each other? As far as I can see, 99% of the drama here is in their conflicts directly with each other and heavy moderation team involvement in it.
From my point of view (as an interested reader and commenter), this latest drama appears to have started partly due to site moderation essentially forcing them into direct conflict with each other via a proposal to adopt norms based on Duncan’s post while Said and others were and continue to be banned from commenting on it.
From this point of view, I don’t see what either of Said or Duncan have done to justify any sort of ban, temporary or not.

JBlack 19 Mar 2023 0:49 UTC
22 points
20
in reply to: DirectedEvolution’s comment on: Dan Luu on “You can only communicate one top priority”
if you have a thousand organisations each pushing in a different cardinal direction in some high-dimensional space, getting backing and making progress based on how important it is to varying numbers of people, that looks a lot like some sort of gradient descent. Maybe this sort of single-issue focus isn’t as inefficient as it might appear?
There are plenty of ways this analogy can break down, and also plenty of ways it can go wrong even within the analogy. A major victory in one direction can easily “overshoot” into a highly sub-optimal state (e.g. revolution), or various factors can consolidate a lot of update power into just two opposed directions (e.g. polarized two-party states).
Plus of course gradient descent is generally based on some error function that can be evaluated precisely and doesn’t change while you’re trying to optimize, neither of which is true in politics, so the analogy is far from perfect.

JBlack 21 Jun 2023 1:53 UTC
20 points
9
on: “Natural is better” is a valuable heuristic
I would agree with the body of this post to an extent, but the main problem is that nearly every instance of the heuristic I see actually in use are in situations where it is inapplicable.
Examples:
- Expecting small fragments of natural systems to be better in an artificial environment such as plants in enclosed office spaces. There are benefits from some plants in general, but a few can produce unnaturally high concentrations of allergens or toxic insecticidal chemicals in such cases, in close proximity to people sharing the same space for an unnaturally large fraction of every day.
- Applying “natural is better” to cases like medicine where typical natural outcomes are to die or suffer lasting injury.
- Continuing to treat some substances as “natural” when they have been artificially processed and concentrated to an extent that never occurs in nature (such as herbal essences).
With that in mind I think that it is a heuristic that is heavily overused, and should be treated with extreme suspicion.

JBlack 13 Jun 2022 3:45 UTC
20 points
0
in reply to: Lone Pine’s comment on: Why all the fuss about recursive self-improvement?
Using the same model:
Should any species that kills thousands of others be allowed to live?
- shouldn’t
Looks pretty straightforward to me.

JBlack 4 May 2023 0:59 UTC
19 points
13
in reply to: Roko’s comment on: Formalizing the “AI x-risk is unlikely because it is ridiculous” argument
Actually the correct conclusion to draw from the graph is that if things continue as they have been, then something utterly unpredictable happens at 2047 because the model breaks there and predicts imaginary numbers for future gross world product.
The graph is not at all an argument for “business as usual”! Bob should be expecting something completely unprecedented to happen in less than 30 years.

JBlack 2 Apr 2023 2:03 UTC
18 points
6
on: AI Safety via Luck
Of course, this does ignore the fact that every human alive is descended from only those luckiest sperm and eggs that made it to fertilization and through gestation. Some of them were even lucky enough to make two or more humans! Hence humans are already evolved to be as lucky as possible, and such a lucky species obviously won’t end up dead due to something as predictable as unaligned AI.

JBlack 30 Apr 2022 2:40 UTC
17 points
on: Increasing Demandingness in EA
I currently have a salary of around $80k/year.
If you believe that I could instead provide the same benefits to civilization by being directly employed by some well-run EA organization as $2.4M/year in donations, then I will happily do this for only $1M/year. Everyone will be very much better off.
Does this sound like a good deal? If not, then how does this square with the N ~300 estimate?

JBlack 31 Jan 2023 23:23 UTC
16 points
8
in reply to: Fer32dwt34r3dfsz’s comment on: Nice Clothes are Good, Actually
For most of my life I thought dressing nicely was a type of bullshit signaling game I didn’t need to play because I was doing Real Work.
So why should you dress nice, even given this challenge? Because dressing nice makes your vibes better and people treat you better and are more willing to accommodate your requests.
The only substantial difference between these two statements is the part that says ”… I didn’t need to play”. It is a bullshit signalling game, but it’s a bullshit signalling game that you do need to play.

JBlack 30 Aug 2022 1:36 UTC
16 points
2
on: LessWrong’s prediction on apocalypse due to AGI (Aug 2022)
According to a “no fire alarm” model for AI risk, your prediction (every survey shows a later date for doomsday) is exactly what should be expected, right up until doomsday happens and there are no more surveys.
In practice, I think there are some leading indicators and numerous members of the community have shortened their timelines over the past year due to more rapid progress than expected in some respects. I don’t know of anyone who has lengthened their timelines over the same period.

JBlack 18 Jan 2023 1:17 UTC
15 points
3
on: ‘Non-profit’ organizations unlikely to exist, as many other types of profit are sought in addition to money—such as social status, ‘mind-share’, political leverage, etc.
Technical terms such as “non-profit organisation” cannot generally be interpreted by just taking the common meanings of the words in English and deriving all sorts of associations from those.
For example, the Australian Tax Office has a classification “Not-For-Profit Organisation” (NFP) which is subdivided into Charities (with further distinctions within that such as Public Benevolent Institutions and Health Promotion Charities) and various other types of NFP such as clubs, community service, professional associations, and cultural societies.
The primary difference between these and other business entities is that these do not distribute profits to owners or members. They may make profits, and people can certainly “profit” in the common sense from their activities, but the profits (in the financial sense) must be retained within the organisation and there are restrictions on how assets may be distributed when being wound up.
Business laws and terminology in many other countries differ in details but are often broadly similar, still referring to financial profits since the legal treatment of those are what distinguishes such organisations from others.

JBlack 4 Feb 2024 10:42 UTC
13 points
5
in reply to: ROM’s comment on: Brute Force Manufactured Consensus is Hiding the Crime of the Century
It doesn’t directly strengthen the lab leak theory: P(emergence at Wuhan & caves distant | leak) is pretty similar to P(emergence at Wuhan & caves nearby | leak).
It does greatly weaken the natural origin theory: P(emergence at Wuhan & caves distant | natural) << P(emergence at Wuhan & caves nearby | natural).
If those are the only credible alternatives, then it greatly increases the posterior odds of the lab leak hypothesis.

JBlack 21 Nov 2022 6:34 UTC
13 points
7
on: Limits to the Controllability of AGI
While I agree with the existence of severe problems with controllable superintelligence, I completely disagree that the arguments in this post prove anything about impossibility. The arguments are primarily about grossly non-central semantics, and not the meaningful substance of these terms.
Furthermore, addressing two extreme examples of a multidimensional space (poorly) and handwaving the rest by saying “trades off between control and safety, but without guaranteeing either” is just lazy writing. No support is given for the assertion that nothing in that space can possibly work, let alone the wider assertion that nothing at all can work.
There are certainly some good points in this post, but I’d much prefer to see the definitive claims weakened to match the support given for them.

JBlack 21 Jun 2022 3:17 UTC
13 points
7
in reply to: mukashi’s comment on: Where I agree and disagree with Eliezer
I absolutely agree. Australia has done substantially better than most other nations regarding COVID from all of economic, health, and lifestyle points of view. The two largest cities did somewhat worse in lifestyle for some periods, but most other places had far fewer and less onerous restrictions than most other countries for nearly 2 years. I personally was very happy to have lived with essentially zero risk of COVID and essentially zero restrictions both personal or economic for more than a year and a half.
A conservative worst-case estimate for costs of an uncontrolled COVID outbreak in Australia was on the order of 300,000 deaths and about $600 billion direct economic loss over 2 years, along with even larger economic impacts from higher-order effects.
We did very much better than that, especially in health outcomes. We had 2,000 deaths up until giving up on elimination in December last year, which was about 0.08 deaths per thousand. Even after giving up on local elimination, we still only have 0.37 per thousand compared with United States at 3.0 per thousand.
Economic losses are also substantially less than US in terms of comparison with the pre-pandemic economy, but the attribution of causes there is much more contentious as with everything to do with economics.

JBlack 6 Jun 2021 6:29 UTC
13 points
on: The Nature of Counterfactuals
A “counterfactual” seems to be just any output of a model given by inputs that were not observed. That is, a counterfactual is conceptually almost identical to a prediction. Even in deterministic universes, being able to make predictions based on incomplete information is likely useful to agents, and ability to handle counterfactuals is basically free if you have anything resembling a predictive model of the world.
If we have a model that Omega’s behaviour requires that anyone choosing box B must receive 10 utility, then our counterfactuals (model outputs) should reflect that. We can of course entertain the idea that Omega doesn’t behave according to such a model, because we have more general models that we can specialize. We must have, or we couldn’t make any sense of text such as “let’s suppose Omega is programmed in such a way...”. That sentence in itself establishes a counterfactual (with a sub-model!), since I have no knowledge in reality of anyone named Omega nor of how they are programmed.
We might also have (for some reason) near-certain knowledge that Amy can’t choose box B, but that wasn’t stated as part of the initial scenario. Finding out that Amy in fact chose box A doesn’t utterly erase the ability to employ a model in which Amy chooses box B, and so asking “what would have happened if Amy chose box B” is still a question with a reasonable answer using our knowledge about Omega. A less satisfactory counterfactual question might be “what would happen if Amy chose box A and didn’t receive 5 utility”.

JBlack 1 Nov 2023 2:20 UTC
12 points
6
on: On Having No Clue
Complexity heuristics possibly work to some extent here. If A is a hypothesis of comparable complexity to B, then you at least have some basis for assigning them comparable prior probabilities.
If A can be conveniently split into A1 and A2, then the child hypotheses are often more complex than A in that they each require some additional property on top of A. This isn’t always the case, but in the exceptional cases it’s likely more useful to consider one or both of the child hypotheses to be “primary” instead of A.
This is all very rough, since complexity is usually only defined up to some constant and is relative to some sort of specification model, but it’s something that isn’t necessarily contingent on observations of the world. If you’re an entity that is considering probabilities and hypotheses at all, the concept of measuring complexity is likely already accessible to you.

JBlack 28 Sep 2023 6:55 UTC
12 points
4
in reply to: Ruby’s comment on: Petrov Day Retrospective, 2023 (re: the most important virtue of Petrov Day & unilaterally promoting it)
How can you honour your word at all if the premise of the link was false for more than half the respondents? There is no action that is consistent with your words.

JBlack 23 Jan 2023 1:44 UTC
12 points
0
in reply to: shminux’s comment on: Quantum Suicide, Decision Theory, and The Multiverse
Absolutely agreed. This problem set-up is broken for exactly this reason.
There is no meaningful question about whether you “should” pay in this scenario, no matter what decision theory you are considering. You either will pay, or you are not in this scenario. This makes almost all of the considerations being discussed irrelevant.
This is why most of the set-ups for such problems talk about “almost perfect” predictors, but then you have to determine exactly how the actual decision is correlated with the prediction, because it makes a great deal of difference to many decision theories.