AnthonyC

Karma: 3,590

AnthonyC 27 Nov 2025 12:36 UTC
6 points
2
in reply to: Kaj_Sotala’s comment on: The Economics of Replacing Call Center Workers With AIs
All possible outcomes, yes! I think the jury question is important, however much it might end up being in some sense silly on the merits. There’s a lot of implementation details that can go wrong.
To add one more thought—done well, there can also be value in ²⁴⁄₇ availability, consistent customer experience, and never getting a busy signal or put on hold.

AnthonyC 26 Nov 2025 0:24 UTC
2 points
0
on: Is Rationalism a Religion
I second the other commentors questioning what, precisely, it is that this distinction is meant to accomplish. And also pointing out the very human and instrumental reasons why movements and causing end up converging on similar structures and dynamics.
To add to that, recall that It is the chief characteristic of the religion of science that it works. Among religions, there are different shades/degrees of how much they reflect the reasons we have concerns about the things we call religions. They come in different shades of gray. Which is exactly why that linked post mentions that “If science is a religion, it is the religion that heals the sick and reveals the secrets of the stars.”
Rationality aims (and claims) to be more in the vein of science in this way, and less in the vein of Christianity.

AnthonyC 26 Nov 2025 0:04 UTC
2 points
0
on: The Enemy Gets The Last Hit
Agreed. Reminds me of something my sensei used to say: It’s useful to strike first, but far better to strike last.

AnthonyC 25 Nov 2025 23:51 UTC
8 points
3
on: The Economics of Replacing Call Center Workers With AIs
Thanks for the analysis, but I think this is only looking at about half the equation.
Does the AI stay on-script in ways that shorten call duration? Or otherwise improve company economics (aka efficiently denying refunds/returns/warranty claims/etc.; or successfully generating conversions to sales, service plans, or whatever; or successfully solving customer problems on the first call)?

AnthonyC 23 Nov 2025 15:05 UTC
2 points
0
in reply to: Mitchell_Porter’s comment on: We won’t solve non-alignment problems by doing research
I do agree with that. I also think it might be worth diverting a rather small percentage of effort towards figuring out what we actually want from and for AI development, in the worlds where that turns out to be possible. At the very least, we can generate some better training data and give models higher-quality feedback.

AnthonyC 22 Nov 2025 3:22 UTC
4 points
0
on: We won’t solve non-alignment problems by doing research
You know, with the funding numbers involved, there’s at least a half dozen companies and a dozen governments that could each unilaterally say, “We’re hiring 10,000 philosophers and other humanities scholars and social scientists to work on this, apply here.” None of them have done so.

AnthonyC 22 Nov 2025 3:18 UTC
2 points
0
on: One King Upon The Chessboard
Worth noting that Dumbledore himself is also not the king on his own chessboard. He deliberately removes himself after setting things up so that, in some sense, he continues to get to make more moves anyway, possibly even up to and including his own future return.

AnthonyC 19 Nov 2025 22:28 UTC
2 points
0
on: Monthly Roundup #36: November 2025
If the job interview was too easy, perhaps you don’t want the job.
I’m curious if this is an experience most people agree with. In my experience, by the time I get an interview at all I’ve already passed the key filters, and usually find the actual interview(s) fairly straightforward. Otherwise companies assume I’m overqualified or decide I fail to tick some unimportant box and I never make it to the “interact with a human” stage.

AnthonyC 19 Nov 2025 21:40 UTC
3 points
1
in reply to: eniteris’s comment on: No One Reads the Original Work
AFAIK it’s the origin of the saying that good fences make good neighbors. A saying the speaker pretty explicitly disagrees with, as a general principle. Yet it’s the saying society remembers.

AnthonyC 19 Nov 2025 2:32 UTC
2 points
0
in reply to: gjm’s comment on: On Writing #2
Reasonable suspicion. Empirically false, in my case.

AnthonyC 19 Nov 2025 2:26 UTC
13 points
5
on: Varieties Of Doom
Strongly upvoted. Great post. I disagree with most of it and would love to read more like it.

AnthonyC 18 Nov 2025 23:02 UTC
2 points
0
on: On Writing #2
What are you trying to say here? Okay, just write that.
Literally /part of every editing session with my boss, my first two years out of grad school.

AnthonyC 18 Nov 2025 14:05 UTC
5 points
2
on: Why is American mass-market tea so terrible?
This is also a good example of something I am seeing as a general trend. In my personal experience, most visible in groceries, rising costs of things like labor and logistics relative to materials create a kind of compression that decreases the proportional cost premium of choosing higher quality. There’s always ways to spend more, and usually ways to find low or average quality cheaper. But I feel like this depends more and more on where and when you shop, rather than the quality of what you buy. I have no hard data at all about this, its just perception.

AnthonyC 18 Nov 2025 13:24 UTC
4 points
0
in reply to: David Gross’s comment on: No One Reads the Original Work
This is true. I think the for OP (and definitely for me) it feels obvious enough that that ‘shouldn’t’ be necessary, but I agree it is. Once you start noticing it, you see it everywhere. But that only happens if you are one of the people who reads the original work.
One of the simpler, though fictional, classic examples is The Road Not Taken, which is very very short, yet even Frost himself complained about how people were misunderstanding it. And even people who do read it seem to keep misunderstanding it. Many people seem to completely gloss over the sentence structures and small-but-critical words in what they read, and end up parsing it incorrectly. Ditto for Mending Wall. I find literary examples useful because they’re less technical (and therefore more accessible) and very easy to confirm.
One example I like is that Einstein wanted to call the theory of relativity the ‘theory of invariants.’ He understood that what matters, as a deep principle of physics, is which quantities and laws are truly fixed. Yes, he showed that some important things people had thought were invariant weren’t, but what mattered were the things that were invariant that people hadn’t realized. Instead even physics teachers talk about ‘paradoxes’ and end up confusing a lot of hapless undergrads, a century later.

AnthonyC 18 Nov 2025 12:51 UTC
2 points
0
on: Don’t grow your org fast
To tell a story, suppose you’re the CTO and you have a new engineering manager underneath you who is a mix of incompetent and vindictive. This manager won’t inform you of the personnel issue here, and the reports unlikely to reach out if the CTO feels inaccessible and the manager is vindictive. It could perhaps be surfaced with skip-level 1-1s. This can happen if the CTO manages two middle managers. What if there are six?
Conversely: This also has implications for how many direct reports the middle manager probably has. If I have skip-level 1 on 1 meetings, or some other upward review process, my response potentially looks very different depending on whether I’m one of 2 or 6 direct reports. 6, and there’s a lot of people saying a lot of things, I’m part of a crowd and can be fairly open and honest. 2, and it’s much harder for me to trust that my boss won’t know who said what about them, if there’s anything that needs to be said that might incur retaliation.
Otherwise, overall agreed with most of this. Many companies and groups scale too quickly because they think it’s what they’re supposed to do, for many reasons. Some others really should scale but can’t or won’t.

AnthonyC 18 Nov 2025 12:34 UTC
4 points
2
in reply to: Adele Lopez’s comment on: Everyday Clean Air
That’s true. For air and surface cleaning the smell is much weaker because the total amounts are much smaller. And it fades very quickly, because it decomposes to saline as it dries. You can safely disinfect a room with people still in it. In some places, businesses do so a few times a day. It can be used as a laundry additive, too. The effect on mammalian cells isn’t zero, but it is much lower than for bacterial, viral, and fungal cells. It’s one of the compounds your body produces internally as a nonspecific antimicrobial.

AnthonyC 15 Nov 2025 16:30 UTC
9 points
7
on: Everyday Clean Air
Is there a reason why hypochlorous acid fogging isn’t part of the conversation as well? Either alone or in combination with UVC. It’s very safe at relevant concentrations, and can sterilize a room in minutes using water, electricity, salt, and vinegar.

AnthonyC 15 Nov 2025 16:15 UTC
2 points
0
on: Some Sun Tsu quotes sound like they’re actually about debates/epistemics
At a sufficient level of abstraction, the fundamental principles of strategy are approximately substrate-independent, I think. They are formless. Void.
what if your enemy also knows themselves and knows you?
IDK what Sun Tzu would say to this, but I would say,
1. No one in practice has actually perfect knowledge, so one side would still have more than the other, and
2. The quote does not say you will win every battle. It says you need not fear the result. It might be that you know the result will be that you will lose, and therefore you should either avoid the battle or surrender from the start. To paraphrase another well-known leader with a challenging strategic situation: If you know you will win, there is no need to fear. If you know you will lose, it’s of no use to fear.

AnthonyC 14 Nov 2025 15:59 UTC
2 points
0
on: AI #142: Common Ground
California’s Attorney General will be hiring an AI expert.
The linked article headline says they’re hiring an AI ‘expert.’ IDK what they’re planning on paying this person, but I doubt it’s enough to hire an expert. An ‘expert’ seems like exactly what they’ll get.

AnthonyC 13 Nov 2025 2:05 UTC
2 points
0
on: The Pope Offers Wisdom
Hmm...