Max Clarke

Karma: 290

Max Clarke Nov 6, 2022, 1:37 AM
5 points
3 ∶ 0
in reply to: Miranda_Zhang’s comment on: Instead of technical research, more people should focus on buying time
I think probably this post should be edited and “focus on low risk interventions first” put in bold in the first sentence and put right next to the pictures. Because the most careless people (possibly like me...) are the ones that will read that and not read the current caveats

Max Clarke Nov 6, 2022, 1:28 AM
2 points
0 ∶ 0
in reply to: Max Clarke’s comment on: Instead of technical research, more people should focus on buying time
An addendum is then:
1. If Buying time interventions are conjunctive (ie. one can cancel out the effect of the others); but technical alignment is disjunctive
2. If the distribution of people performing both kinds of intervention is mostly towards the lower end of thoughtfulness/competence, (which we should imo expect)
Then technical alignment is a better recommendation for most people.

In fact it suggests that the graph in the post should be reversed (but the axis at the bottom should be social competence rather than technical competence)

Max Clarke Nov 6, 2022, 1:20 AM
8 points
3 ∶ 0
in reply to: Miranda_Zhang’s comment on: Instead of technical research, more people should focus on buying time
Rob puts it well in his comment as “social coordination”. If someone tries “buying time” interventions and fails, I think that because of largely social effects, poorly done “buying time” interventions have potential to both fail at buying time and preclude further coordination with mainstream ML. So net negative effect.

On the other hand, technical alignment does not have this risk.

I agree that technical alignment has the risk of accelerating timelines though.

But if someone tries technical alignment and fails to produce results, that has no impact compared to a counterfactual where they just did web dev or something.

My reference point here is the anecdotal disdain (from Twitter, YouTube, can dm if you want) some in the ML community have for anyone who they perceive to be slowing them down.

Max Clarke Nov 6, 2022, 1:10 AM
2 points
1 ∶ 0
in reply to: RobertM’s comment on: Instead of technical research, more people should focus on buying time
That’s a reasonable point—the way this would reflect in the above graph is then wider uncertainty around technical alignment at the high end of researcher ability

Max Clarke Nov 6, 2022, 1:04 AM
5 points
4 ∶ 0
on: Instead of technical research, more people should focus on buying time
I would push back a little, the main thing is that buying time interventions obviously have significant sign uncertainty. Eg. your graph on median researcher “buying time” vs technical alignment, I think should have very wide error at the low end of “buying time”, going significantly below 0 within the 95% confidence interval. Technical alignment is lots less risky to that extent.

Max Clarke Nov 6, 2022, 12:18 AM
1 point
0 ∶ 0
on: ML Safety Scholars Summer 2022 Retrospective
Out of interest, were you considering students working together and thus submitting similar work as being plagiarism? Or was it more just a lot of cases of some students fully copy/pasting another’s work against their wishes?

Sounds like it was a very successful program!

Max Clarke Nov 5, 2022, 10:48 PM
1 point
0 ∶ 0
on: Draft Amnesty Day: an event we might run on the Forum
I think this should be acommpanied by a message/prompt in the comment text field that tells people that this post was a draft and to be err in favour of not giving any negative feedback

Max Clarke Oct 31, 2022, 9:54 AM
2 points
1 ∶ 0
on: Using artificial intelligence (machine vision) to increase the effectiveness of human-wildlife conflict mitigations could benefit WAW
Sorry for a negative comment, but I think that all of these interventions fail to really address wild animal suffering, and that that is pretty clear already. This is simply due to the fact that pretty much all interventions on WAW have only a temporary positive effect, or worse are zeroed out completely, by the malthusian trap.
- wind farm deaths: probably not worse than death by predation or starvation. Also other people’s hobby horse already.
- culling: not worse than predation or starvation. IMO often net positive, keeps populations in growth mode. (Of course, that depends on the suffering of insects though)
- deterrence: temporary negative effect as some birds that were previously eating the crops starve, followed by no effect so long as the deterrence
(Addressing those “probably”s is probably useful)

Ok, well, that’s not the point. As you say in the conclusion:

“[...] Additionally, intervening in HWC is seen as acceptable; [...]”

I think we should be clear that that’s the core proposal here, and thus also build on that more. The artificial intelligence part is just a potentially useful tool. (Although now that I say this, that is potentially a hard post to write in the open? Maybe you tactically alluded to this but not actually say it? And if so and I missed it then I’ll delete this comment)

The unique thing AI can potentially is as you say, scale, but not scaling pest control & deterrence—rather future interventions that make progress on fixing the wild animal malthusian trap and ecosystem effects (ie. predation). Those sort of interventions are unavoidably going to sound like sci fi (😂 like a VR headset and free condoms for every rat 😂), but you can ground it as much as you can.
What links here?
- Neil_Dullaghan🔹 's comment on Using artificial intelligence (machine vision) to increase the effectiveness of human-wildlife conflict mitigations could benefit WAW by Rethink Priorities (Nov 9, 2022, 7:42 PM; 14 points)

Max Clarke Oct 25, 2022, 2:18 PM
1 point
0 ∶ 0
in reply to: Emrik’s comment on: My experience experimenting with a bunch of antidepressants I’d never heard of
No I mean the one I replied to, your sub-comment

Max Clarke Oct 23, 2022, 10:11 AM
2 points
1 ∶ 0
in reply to: Emrik’s comment on: My experience experimenting with a bunch of antidepressants I’d never heard of
Wow people really downvoted it. I just ignored it, in general I don’t like to downvote people who are talking about their poor mental health 🤷‍♂️

Max Clarke Oct 23, 2022, 10:08 AM
2 points
0 ∶ 1
in reply to: Emrik’s comment on: My experience experimenting with a bunch of antidepressants I’d never heard of
Not sure why people were using the main downvote button on this one, and not just the disagree downvote.

Max Clarke Oct 19, 2022, 8:23 PM
1 point
0 ∶ 0
in reply to: Emrik’s comment on: My experience experimenting with a bunch of antidepressants I’d never heard of
Where’s the part of the post you are responding to?

Max Clarke Oct 8, 2022, 8:52 AM
5 points
0 ∶ 0
on: Buddha Was An Effective Altruist
I have been attending a Secular Buddhist group for a couple of years and I have also seen this similarity.
My main idea about how to link EA and Buddhism is as follows:
1. People often become involved in Buddhism for the meditation and mindfulness—to learn about their own minds, change their experience of the world, and have a better life.
2. There are a lot of places where having greater impact as an EA is helped by developing positive motivation and different desires, through practicing meditation and mindfulness.
  1. Eg. For example, the idea of “purchasing your fuzzies and utilons separately” is a good one, but even better is to “align your fuzzies with the utilons”, where you can learn to feel strongly positive about things that might have been boring or impersonal before.
3. An “EA Buddhism” might focus on changing your mindset first, and the actual impact second. For example, it might be a regular Buddhist meditation group, but at some point when peopl have a question along the lines of “Where am I aiming?” the answer can be along the lines of “to become someone for whom doing efficient altruism is aligned with doing efficient hedonism”
4. Problems:
  1. In a buddhist context rather than in the EA context, the answer to “Where am I aiming?” is probably hard to explore too deeply, because the people involved won’t all be intellectual and/or have good epistemics.
  2. Also in a secular buddhist context, emphasizing that one do the “maximum amount” of altruism is probably not tenable. The meaning of EA here should definitely be “efficient altruism” and not “effective altruism”.

Max Clarke Oct 5, 2022, 8:18 AM
12 points
0 ∶ 0
on: Can we actually confirm that a vegan diet is compatible with high intellectual output?
No idea how to go about finding information on this, but by my personal priors I would weight various kinds of evidence as follows:
- ~0 weight on any anecdotal evidence,
- low weight on studies and clinical trials,
- moderate weight on arguments by biological plausibility.
Being related to diet, my prior is that people are usually over thinking it. However I have always agreed that it seems unlikely that a fully vegan diet has no nutritional downsides without supplementation.
I’ve done a cursory search, just wikipedia, here are my thoughts on the biological plausibility of deficiencies among vegans in your two suggestions, creatine and choline:
- creatine
  - is only found in animal tissue
  - is absorbed in digestion
  - is also metabolized from amino acids found in plants
  - ->
  - seems like it could plausibly be something that vegans are deficient in
- choline
  - is an essential dietary component, not being metabolized sufficiently in enough quantity
  - deficiency is rare in humans
  - is found in the same quantity per gram in wheat as in chicken
  - ->
  - seems unlikely that it is something that vegans are deficient in

Max Clarke Oct 5, 2022, 8:01 AM
1 point
0 ∶ 0
on: Can we actually confirm that a vegan diet is compatible with high intellectual output?
Seems like a Q more suited for LW?

Max Clarke Sep 28, 2022, 3:37 AM
3 points
0 ∶ 0
on: Max Clarke’s Shortform
A defense of the inner ring, excerpts from the original.
...
I must now make a distinction. I am not going to say that the existence of Inner Rings is an Evil. It is certainly unavoidable. There must be confidential discussions: and it is not only a bad thing, it is (in itself) a good thing, that personal friendship should grow up between those who work together. And it is perhaps impossible that the official hierarchy of any organisation should coincide with its actual workings. If the wisest and most energetic people held the highest spots, it might coincide; since they often do not, there must be people in high positions who are really deadweights and people in lower positions who are more important than their rank and seniority would lead you to suppose. It is necessary: and perhaps it is not a necessary evil. But the desire which draws us into Inner Rings is another matter. A thing may be morally neutral and yet the desire for that thing may be dangerous.
...
The quest of the Inner Ring will break your hearts unless you break it. But if you break it, a surprising result will follow. If in your working hours you make the work your end, you will presently find yourself all unawares inside the only circle in your profession that really matters. You will be one of the sound craftsmen, and other sound craftsmen will know it. This group of craftsmen will by no means coincide with the Inner Ring or the Important People or the People in the Know. It will not shape that professional policy or work up that professional influence which fights for the profession as a whole against the public: nor will it lead to those periodic scandals and crises which the Inner Ring produces. But it will do those things which that profession exists to do and will in the long run be responsible for all the respect which that profession in fact enjoys and which the speeches and advertisements cannot maintain.
...
— C.S. Lewis, “The Inner Ring”
Fill your heart not with acidic desires
to impress,
or to accomplish;
to go to EA Global,
or to win the approval of “the core EAs”.
And fill it instead with a burning resolve
to help all beings live good lives
now and forever.

Then, to the extent
that EA inner rings succeed at their purpose
you will find yourself in one —
which you once wanted terminally
but now, instrumentally.
An EA inner ring is just a resource
for the betterment of others.
It is utilized, but not exploited
enjoyed, but not pursued.
It is comfortably inhabited —
but left empty at the start of every day
as the sun rises over the early morning frost.

Max Clarke’s Quick takes

Max ClarkeSep 28, 2022, 3:37 AM

3 points

1 comment EA link

Max Clarke Sep 25, 2022, 11:38 PM
1 point
1 ∶ 0
in reply to: t46’s comment on: EA architect: Building an EA hub in the Great Lakes region, Ohio
Hey, appreciate your response. Perhaps we should discuss the meaning of the word “hub” here? To me, it is about 1) Having enough EAs to establish beneficial network effects, and 2) to have a reason why the EAs living there aren’t constantly incentivised to move elsewhere (which also means they can live and and work there if they choose)

I think that your value proposition of a beautiful, cheap location for remote work is a great reason for a hub! This fulfills condition 2). Then, having enough people fulfills 1).

However, network effects cause increasing returns to moving there. This means if there are two hubs (of the same type—here “good cheap remote hub”) that a person might move to, they will all else equal choose to move to the larger one. This means that there will tend to be only one hub of each kind—a lock in effect.

Given that about 30% of all EAs are from the US, and the ~2 existing hubs there are quite expensive, I think your hub has a good chance of succeeding!

However, I think that would result in the above lock-in effect. If so, there might be as many people who live in your northeast US hub (~30% of EAs outside the US and UK) who would like to move there, but can’t (easily) because of US immigration. Having a hub that offered the same value proposition ( condition 2) ) but in a different jurisdiction would have been strictly better.

On the other hand, if you can change the US immigration rules (which are btw definitely bad for the US!!), please do!

Max Clarke Sep 18, 2022, 11:49 AM
8 points
3 ∶ 2
on: EA architect: Building an EA hub in the Great Lakes region, Ohio
Hi!

I have to say I strongly disagree with this idea, for one particular reason. If we successfully establish a new hub with cheap living costs and beautiful nature, it MUST be outside the USA. The USA is notoriously hard to immigrate into from most countries!

It is unfortunate that we already have one hub (SF Bay Area / Berkeley) in the USA, although I definitely am OK with D.C. becoming a hub. However, I’d ask any Americans who want to be in an EA hub, but don’t want to be in those two places, to go to someone else’s hub (Mexico City, Cape Town), or if still wanting to set one up, to do so in a jurisdiction with permissive immigration.

Max Clarke 10 Aug 2022 3:01 UTC
2 points
0 ∶ 0
in reply to: 𝕮𝖎𝖓𝖊𝖗𝖆’s comment on: Are “Bad People” Really Unwelcome in EA?
Yeah the example above with choosing to not get promoted or not recieve funding is a more realistic scenario.

I agree these situations are somewhat rare in practice.

Re. AI Safety, my point was that these situations are especially rare there (among people who agree it’s a problem, which is about states of knowledge anyway, not about goals)

Thanks for this post, I think it’s a good discussion.