svara

"tepid legality"

It's the kind of thing where it's sort of expected that you wouldn't know, right?

I think we don't really understand why deep learning works as well as it does, the thinking around that is, as far as I can tell, mostly a collection of empirical observations.

A fundamental theory of learning that can be used to predict optimal network architectures might enable smaller models that consume less energy.

I want to agree with this, but I have a hard time seeing how it can be done.

Tao is speaking of a very particular kind of mathematics, that done out of pure curiosity.

But maths, even at the highest levels, often finds applications sooner or later.

It will be economically impossible to justify boycotting correct mathematics that no humans understand on grounds only of purity.

This may happen very soon: one of the obvious applications of novel mathematical results is in building stronger AI models.

In animal welfare laws the principle is typically the capacity for suffering.

The argument goes that livestock have a capacity for suffering, but killing them for meat without causing them suffering is ethical.

You can disagree with the position, or with its implementation in practice, but it's a consistent position in principle.

That's too simplistic, since it is in fact very common to be concerned about animal welfare even among people who do eat meat.

> Apple has never had a strategic problem meeting the demand for Apple computers.

The business that had those problems didn't become Apple.

Absolutely. It's the most visible in services shrinkflation.

Amazing, let's please do Japan and Taiwan next.

I don't think we'd be okay with that if it took a few years longer, right?

I thought the Bengio blog post posted here the other day answered that question quite well (ignoring your timeline).

The argument is that in training for the capability to achieve certain goals, you may inadvertently also train for secondary instrumental goals you did not intend, such as aggressive behavior or deception.

I'm actually no sure I buy that this is particularly likely, or necessary, but I found that take reasonable and worth considering.

I'm a bit worried that this line of reasoning is assuming its conclusion, in stating that mathematics is for human understanding.

This is not something that's just true on its own; it's on us as a species to make sure that it remains that way, for the sake of our dignity.

However, this position will be extremely difficult to defend, against the economic value of not caring.

This has a clear answer in systems theory; the ability to bang out code was never the limiting factor for Google et al.; large organizations are limited by coordination costs.

It's the small teams you need to pay attention to, the organizations that really were limited by engineering capacity. These are by definition also less visible - for now.

Yes, but it's also building other skills. There's no friction to learning something new anymore, and that's really quite amazing.

E.g. I've gotten pretty good at git thanks to AI because it forced me into more complex workflows. I still can't remember the commands but I really grok the concepts and terminology on a far deeper level than before, where I made a conscious effort to stay on the happy path lest I get into some odd state I don't understand. Pretty sure those "odd states" would seem quite clear to me now.

> We're reading more words than ever, but instead of coming from books, they're coming from digital media. We're losing that same muscle for long-form reading and deep focus (...)

Yes, true, but that's also very much a thing you can do something about by, well: reading books.

It's a technique we must preserve for the good of our culture, but, also, in a society full of people who have forgotten how to, an individual superpower.

A bit ironic that this is written in idiomatic Claudese.

I'm probably going to be going against the grain here, but I think it's not as bad as it looks at first.

I was similarly frustrated a few months ago, but have noticed I've started to learn the idiom.

Its use of "dense jargon" and "stilted metaphor" is actually surprisingly consistent - it's speaking its own dialect, and you get used to it.

After a while it gets much easier to read and even becomes somewhat efficient, I think, since the odd metaphors it uses often have a precise meaning in Opus-ese (Fable speaks a really similar dialect).

AI company leadership are telling everyone, and that's paraphrasing only lightly, that there's a chance AI will destroy humanity, and will probably cause mass unemployment, so we need to put datacenters everywhere.

So why do people not like datacenters? Go figure.

Absolutely not. I've spent nearly every summer in the last 30 years in that region and it never was like this; it also wasn't that warm.

Occasional fires, sure. But very clearly nothing approaching the current scale and frequency.

The last really big one in that region was in 2022; and there are currently not one but 2 parallel fires in the same forest.

People have not suddenly become deranged or simple minded.

Two of the large recent fires in the Landes region, including the most recent, have started from poorly maintained, spontaneously igniting cars. Those also have not suddenly started appearing out of nowhere.

The common thread is very dry forests, it just takes a spark for disaster.

> Le sujet, c’est l’aridification du milieu. Pour revenir à cet incendie, ce n’est pas un problème de culture, mais de climat.

A bit of a distinction without a difference. The climate is not changing back, it's the new normal. In that climate, this particular kind of agriculture (huge pine monocultures) has become a problem.

It breaks my heart to say this, I've spent nearly all my childhood summers in my family's summer house down there. This year, I left just a few days before the evacuation order would have come.

It's going to be burning every year now and if we want to be safe there, the forest can't stay the same.

The pine forest is the agriculture.

You're not wrong, but there are levels to understanding.

For example, when studying maths or physics, it is very common to feel like you've understood everything, until you get to the exercises, and need to apply that understanding, and it's only then that you consolidate the knowledge and begin to truly understand in depth.

I like AI enhanced coding, but I do sometimes worry that we're not getting enough of that depth anymore.

Somehow all the with/without cookbook design examples look vibe coded on both sides to me.

I'm not enough of a designer to be able to point exactly at what makes it so, but Claude does seem to have a somewhat limited repertoire of styles.

Maybe if you could point more precisely at the required changes, you could discourage it?

OK, but having read that section you'll see how

> just writes off any improvement

Is not what that section contains?

There's a more nuanced point about capabilities improving in some ways but not others (ability to one-shot apps vs. maintenance of existing code over time).

Genuinely happy to hear you're successful here! But, why would you expect there to be no drawback to not knowing the local language when moving to a foreign country?

You are (as am I) in the 30% of Germans (40% in major cities) with a Migrationshintergrund.

At that point, it barely makes sense to call that a minority, it's just normalcy. If you find yourself in a pocket of unusual backwardness where it feels otherwise, you should probably leave.

I pass as German based on looks, but my name is weird and my wife doesn't look or sound German at all. I don't think her or I have ever noticed any adverse consequences from that.

If your German is good, you can just act and feel like you belong here and no one will challenge that.

The people saying they're having trouble getting by with just English though are weird to me. What did they expect? Different countries are different, that's sort of the point.

I do actually agree that Germany isn't the best country when you're looking for economic opportunity, but that isn't really what people are optimizing for here. You might disagree with this, but it's mostly not directed against immigrants.

Regarding your political points: Ironically, they sound very German to me. Yours is a standard left of center critique in German politics. The countries that have a long history of being targets for immigration largely don't work that way, probably because extensive social safety nets are bad for the acceptance of recent immigrants by locals.

Thanks for actually giving a great answer to this versus the other knee-jerk comments! I will definitely look into this more.

I think it's important to note that the cost of bringing a drug to market has increased a lot (about threefold) since 1990, so if the case studies are that old they might not generalize.

It's not black and white- in pharma there are notable sources for investment that are not private, e.g. for orphan diseases that would otherwise not see enough investment due to lack of profitability.

But here's a mechanism to mobilize private capital for those diseases that are. Again, the investment needed boogles the mind, around 10B currently, for a single drug.

You're asking someone to pay that money and allowing others to reap the rewards. Why wouldn't everyone want to be the one who didn't pay the money?

Generics exist for drugs that usually someone did hold a patent for that has since expired and they're much less profitable than patented drugs.

Not sure how that address the point. Again, the investment to get a drug to market is gigantic and you're saying that someone should pay for that and someone else should get the reward. It really doesn't work that way. Everyone would want to be the one who didn't pay.

No, only a small part of this is publicly funded.

You spend billions to get a drug from concept to approval - and then once you've invested all that money, someone else can just sell it too, free loading on all the studies you ran? Why would anyone invest in drug studies?

I need a bit more depth and detail to believe that this doesn't destroy the pharma industry.

What would the empirical evidence even look like? It's not like the modern pharma industry existed before patents.