Shrike
4hr ago
Right, "even the month long preview period won't tell us what these models can and cannot do" is also true for humans. We cannot tell if a human might design a deadly new virus. To the extent that AIs are a problem, they are a problem in the same direction as widespread literacy, accessible libraries of books, and the internet: the broadening of knowledge also broadens the circle of people who might abuse this knowledge.
The best solution is pretty clearly physically securing dangerous materials.
Shrike
4hr ago
Since splitting from the Catholic church, as far as I know. But in Anglicanism it is not mandatory ("all may, some should, none must"), so perhaps saying it was practiced in a similar way was misleading. From what I understand the form of confession to the priest is very similar, though.
Shrike
7hr ago
From my perspective, the biggest difference is Catholics have the Sacrament of Reconciliation, which provides definitive ways to know a sin has been forgiven. Whereas Protestants have to hope they've been forgiven, and that ambiguity doesn't seem like a positive to me.
I don't think this is really right at all.
Firstly, confession is not unique to Catholics; Anglicans practice it in a much similar way and Protestants in different ways.
Secondly, while it is difficult to speak for all Protestants since they have differing views of soteriology, it's not particularly different to find denominations that teach that you can definitively know that your sins have been forgiven. In fact I think the only denominations that wouldn't teach that (off the top of my head) would be, ironically, the hardcore (for lack of a better term, though there probably is one) Calvinist and Arminian fringes, which might insist, respectively, that one cannot know one is elect and that one loses their salvation through sinning.
Obviously plenty of Protestants worry about their salvation and the forgiveness of their sins, regardless of what their denomination may teach, but I don't think that Catholics are much different in this regard.
Shrike
1d ago
The pun in Hail Mary is a reference to the prayer ("Hail Mary, full of Grace")
Shrike
1d ago
The US, burdened by its gigantic array of global missions, cottoned on just a few years ago.
No, the US has been concerned about this for a very long time, publicly so since the Obama administration.
Now the newcomer is going to out-deny the Chinese in China's home waters, from the other side of the world?
If China could succeed at their goals by keeping their fleet in their home waters, why would they need a blue water navy at all? (Or really a navy at all period?)
The answer is that a blue water navy, at a minimum, lets them credibly threaten things they cannot otherwise credibly threaten. But to do at least some of those things, they will need to leave their home waters.
Ports are more easily mined and bombed than before.
Either this is true and the US can credibly threaten to conduct area denial in China's home waters, or it isn't and it's a nuisance at best to the US coalition, pick one.
The whole of Taiwan is within Chinese SAM range, whatever issues China has in the killchain are going to be dwarfed by the struggles the US will be having.
In real life, SAM ranges aren't little circles on a map that let you just delete anything in that range. Maneuvering fighter aircraft over Taiwan can probably kinematically defeat Chinese ground-based SAMS, but even if they can't Taiwan's mountains makes terrain masking trivial. There's a reason China is investing in an air force.
The cheap drones destroy or threaten to destroy military targets, ports, radars, fuel storage, power plants... Taiwan capitulates.
If China intends this strategy, why are they bothering to procure a credible blue-water navy, complete with carriers? Actually, why bother with drones at all? A few nuclear weapons would also achieve this end, probably more cheaply.
US tactics will always be shaped by what their enemy does. If their enemy is buying a navy instead of cheap drones (which as you've said yourself there's no evidence either of us are aware of that China is pumping out massive amounts of cheap drones for their own military, and there is definitely evidence they are doing this with ships) then of course the US is going to optimize to defeat that threat. Of course, the US is also ramping up their counter-drone efforts as well, but you go to war with what you have, and that is as true in China's case as everyone else's.
Either your plan isn't as good as you think it is, or you're smarter than the Chinese, who are wasting their money building aircraft carriers when they could be pumping out Shaheds, or some combination of the two. Either way, the US will need to deal with a large conventional surface fleet fighting China.
The US has a grand total of 4 minesweepers deployed in Japan. I do not understand how 4 Avenger minesweepers are supposed to get close to a Taiwanese and Korean port for an extended minesweeping mission when US doctrine is 'stay as far from Chinese shores as possible' with well-defended carrier groups. The Japanese minesweepers face similar issues.
The Avengers probably aren't supposed to do that, since most US minesweeping these days seems to be done with helicopters and USVs (which I think are pretty small). Note how the US swept the Gulf recently without redeploying minesweeping ships there.
You talk about various US lasers but why can't we actually see them doing anything?
I linked a US laser system being demonstrated to you less than six months ago.
If US lasers actually worked (to an extent still not reaching the demands of a war with China), the 5th fleet's base wouldn't be ruins. Soldiers wouldn't have been hiding in hotels. There wouldn't be a drought of Patriots and THAAD because the cheap lasers would be shooting down these drones and missiles instead.
Your argument is that if US lasers – which so far have been deployed in limited numbers, mostly as sea-based point-defense weapons – actually worked, then they would be acting as area-defense weapons to protect mostly inland targets?
It's true that the US has not procured laser weapons on a large scale yet. It's also true that it's logically possible for US laser weapons to be effective but not yet deployed in sufficient numbers. It is additionally true that it's logically possible for US laser weapons to be extremely effective but not perfectly so, so that assets protected by them could still be damaged for any number of reasons despite a high rate of success. And, finally, it's extremely true that point defense weapons mounted on ships are not area defense weapons. Thus, even if the US had successfully used laser weapons against Iran (which quite possibly we haven't, the DoW has been coy about it), we would still expect to see the bad effects of Iranian drones on US installations ashore.
The US navy struggled with Yemen's missiles. They struggle with Iran's missiles and drones. China is a much bigger threat.
Yemen never hit a US ship, and Iran killed about a dozen American servicemen despite firing more than 1,500 ballistic missiles and around 4,000 drones. The US "struggles" because their standards are extremely high, but despite some high-profile (and bad!) equipment losses, they are actually extremely good at preventing service casualties to Iranian attacks.
I can only imagine you're implying a nuclear exchange, which might've made sense 25 years ago with Russia sidelined. Too late now.
Nah, I mean more blockading China which can be done at standoff distance without getting anywhere near most of China's anti-access denial weapons. Since about half of China's trade and nearly all of its physical goods are moved by sea, this plus the contraction from the loss of oil (about 20% of China's energy) would kneecap the Chinese economy. The US could sustain a sanctions-and-surface-raiding program for a very long time, probably indefinitely, and use that window to shift worldwide industrial dependency to it and its allies. It's true that eventually China would route around the damage, but it would be both systematic and severe and quite plausibly permanent in the sense that China might never recover entirely. China's options if the US decides to do this are "suck it up," "try to use soft-power to change the facts on the ground," or "sally from Chinese territorial waters with that blue water navy we talked about" which would almost certainly end poorly for them.
Would the US do this? I don't know. I don't know if China will try to invade Taiwan. But I am pretty sure the US can do this.
Shrike
1d ago
Iran did in fact attack Turkey, they just did not do so successfully.
Shrike
1d ago
Fair enough distinction, although it sounds like things are already pretty bad over there after just a short blockade.
Shrike
3d ago
I don't think this is strictly speaking true. The US didn't need regime change in Cuba for the Caribbean to become an area of much less concern to it, for instance: sufficiently degraded capabilities suffice.
I do think this (or at least the specific example of Cuba) suggests that long-term sanctions that stick against Iran would probably be necessary in the alternative to keeping a major force presence, and I'm not sure if Iran could be as easily sanctioned as Cuba.
Shrike
3d ago
So the US plans to shut down the strait with 1100 km range F-35Cs firing 370 km range anti-ship missiles. Plus B-21s and subs.
And land-based mines and surface-to-surface cruise/ballistic/hypersonic missiles.
LUCAS or 'cheap hypersonics' are fields China's going to be ahead in.
The United States actually has fielded LUCAS in combat. China has not. While I know Chinese companies have put together Shahed clones, it's unclear to me if the Chinese military has even procured any of them, while the US is actively procuring and integrating LUCAS into its order of battle.
And while China likely has a current edge in air-launched hypersonic weapons, the US has a "cheap" hypersonic missile (Blackbeard) in flight testing now, and it seems quite plausible to me that the US is ahead of China in relevant areas (particularly sensor and propulsion technology) that will allow them to bend the cost curve down in their favor.
No matter how many LUCAS's America makes, China will make a lot more more.
Again, this is not the operative dynamic that matters here.
If the West Pacific becomes an impassable no man's land, then Taiwan and South Korea will capitulate eventually due to their total reliance on seaborne imports, lack of domestic energy/food supply or import-substitution capability. Japan is also in some danger. Shutting down the strait only removes China's preferred victory condition of a quick win, it's insufficient.
Setting aside that, first of all, it's a violation of international law to shut off food supplies, and that it would likely be pretty foolish for China to pick a fight with South Korea too, how has this worked for everyone else that has tried it? Iran has no shortage of ballistic missiles and drones, but they have not been able to shut down the strait of Hormuz, which is a much more constrained location than the Philippine Sea and the Sea of Japan. Ukraine, as far as I can tell, has not managed to shut down the Black Sea for Russia (or vice-versa). The Houthis had mixed results. It did not work for Germany in World War One or Two. I think the US versus Japan was probably the most successful submarine warfare campaign in the world, and the US still did not defeat Japan through starvation.
Secondly, how are you supposed to turn the entire West Pacific into a no-man's land? You've got a few options:
Mines – but there's not enough mines to do this to the entire West Pacific. You can use mines to shut down chokepoints, not the entire ocean. And the US has recently demonstrated the ability to clear mines pretty quickly (although Iran did not deploy large numbers in the strait at all – something like a dozen real mines and a couple hundred decoys). Might work against Taiwan, or temporarily against ports elsewhere. You're not going to starve out Japan or South Korea with mines, or turn the entire West Pacific into a no-go zone.
I don't think China has enough good submarines to shut down traffic to South Korea, Japan, and Taiwan.
You can use radar to try to detect anything that moves in the entire West Pacific, but ground-based radars don't have the range (barring OTH radars, which from what I understand are finicky and may not have the resolution to complete a kill chain). Air-based radars have better range, but AWACS flying over Chinese soil probably don't have the radar range to detect and track ships traveling to South Korea, let alone Japan, just due to the curvature of the earth. AWACS are also very big, slow, juicy targets, so pushing them out further is very dangerous, and it's possible to jam them.
Satellites are the best option, but they will likely be slated for deception, destruction, or defeat early on in a conflict. Satellites also have limitations due to weather or inclination; the Soviet Union had RORSAT deployed during the Cold War, but that did not stop them from losing track of NATO carrier groups operating near their shores in the 1980s.
High-speed reconnaissance aircraft are an interesting possibility that China has invested in, but you they are not invincible.
The US is the side that needs some kind of laser to shoot down Shaheds, not China.
I guess it's good that the US has this, then!
The US victory condition is preventing Taiwanese and Korean subjugation.
Or perhaps, if China actually decides to conduct aggressive action, the United States will decide that its victory condition is crushing China, and would gladly trade Taiwan and Korea to do so.
Shrike
3d ago
Given that AI service providers are essentially charging users for content, locking users into arguing with models/a certain base rate of mistakes/verbosity (bumping up use rates) might be a viable revenue strategy, with the mitigating factors being performance competition from other providers, the need not to totally alienate users, and the fact that such a strategy might backfire if mismanaged since they are themselves renting compute.
If incentives were structured differently (for instance, if AI service providers were paid a fixed fee per task) there would (I think) be stronger incentives for models to be extremely efficient. As it is, right now it's in the interest of AI service providers for their models to be compute-heavy – sure, it is costly, but it lets them run back to their investors and spin a tale of how much demand there is for the product.
I'm not sure if this explains the verbosity of the models but I've certainly wondered if it plays a role.
Shrike
3d ago
I don't think Iran was done as a 5D chess move to distract from negative immigration news bites, but I wouldn't be surprised if, now that it is kicked off, the Trump administration (or at least some factions within it) aren't (contra what you would expect from Reading The News) desperate to end it in any rush exactly because it is generating narrative top cover as you describe.
I also do think that the blockade will continue to pay off over the long run and the admin is probably in less of a rush to "win the war" than one might imagine for that reason as well.
Shrike
4d ago
In some sense it may be better for Trump['s project] to be unpopular due to the war than something like immigration enforcement. If you think that some degree of unpopularity was baked in, the unpopularity of the war is something that the administration can to some degree steer (and also that future Republican nominees can more easily disavow).
Shrike
4d ago
The Iran-Oman agreement bans US warships.
If you are reading @coffee_enjoyer's comment without clicking through on the links, you might have been under the impression that this "authoritative information" was based on a done deal, the details attested to by both sides. In fact it is based on an interview with a member of Iran's foreign ministry about a deal that has not been signed.
South Korean decapitation strike on the absolute monarch of an insecure nuclear power seems like a very risky ploy.
Well, tell it to the South Koreans. I don't think that North Korea and China are joined at the hip the way you seem to think they are (ditto for South Korea and the United States), nor do I think sending some guys to Ukraine, at roughly zero risk to the North Korean regime, is any indicator of how they will behave in a potentially existential situation.
If swarms of relatively cheap Shaheds and FPV drones are the way of the future, then it plays into China's biggest strengths in industrial scale
Now you're just taunting me. You know that FPVs are not the future of naval warfare.
The Chinese economy grows faster than the US, they're breaking into more and more markets globally, everyone started bitching about overproduction.
I wouldn't be so confident about the Chinese economy if I was you – but this is a positive reason why China may decide to sit around and do nothing. And yet if they do that, I think there is a good chance that the US is actually able to procure enough missiles, mines and torpedoes to make shutting down the strait a ~foregone conclusion. China can try to build more ships than LRASMs, more interceptors than the cheap hypersonic and LUCAS cruise missiles the US is producing, but they will be on the wrong side of the cost curve in doing so.
Even their geography has its advantages, they're right next to the places they'd be fighting, whereas America is on the other side of the world.
Sort of - in some ways, the US has the best of both worlds here. Japan, South Korea, and the Philippines host US bases, and if a war kicks off, the entire Chinese navy will be ranged by missiles based there, while portions of the US Navy will be safely out of range of conventional Chinese missiles. China does have strategic depth, but it is mostly for their land and air forces.
Shrike
4d ago
(Also I conceptualize Korea getting dragged into a US-China war one way or another, so there would be a major ground element where shell production is significant)
It seems fairly plausible that South Korea will get dragged in to some degree or another (possibly below the level of kinetic action), but it's unclear to me what advantage it would be to North Korea to get involved kinetically. South Korean doctrine at the moment is built around "KILL KIM JONG UN" if the balloon goes up and I think the rational move from Kim is to try to extract concessions from both sides and basically nothing else. I am certainly open to alternative takes, but it seems common in Western thinking to assume that if China suggests North Korea jump, the North Koreans will ask "how high?" because they wear the designated villain hat, and I don't think actual North Korean decision-makers have any emotional attachment to being committed to being part of the Axis of Evil. I do think they have emotional attachment to not ending up as paste at the pointy end of a South Korean ballistic missile.
You raised some qualitative edges in weapons systems but the Chinese already have similar systems in development. If America can make a missile, so can China but China will have more engineers to juice them up and more production capacity to produce them at scale, just by virtue of this huge industrial base and economies of scale. Unless the missile-edge enables the US to win in a pre-empting war then what good is it, it's only moving the goalposts. China can then jog closer and score, they're faster at running than the US is at moving goalposts (with the exception of space and AI).
Again, you're sort of missing the main point here. The USA does not need to match China to win; it already has the "offsetting advantage" which is the status quo (and geography). China needs to overmatch the US of A because it is starting from a geographically disadvantaged position. I think that there is a good chance that China does nothing while the US painfully rectifies its munitions production and stockpiles issues over the next 5 - 10 years precisely because the risks of acting even when China is at the apex of its power relative to its potential enemies are still extremely high, and Xi does not want to gamble with his empire.
Shrike
4d ago
Thank you! I've liked RUSI's stuff on Ukraine that I've reviewed, I'll check these out.
Not sure if the Carnegie links are different, but it looks like the second one is missing.
Shrike
5d ago
I haven't really heard that perspective outside of the red flag propoganda-dripped areas, and I know that's not what you base your opinion on.
There was a report by the Swiss Centre for Military History and Perspective Studies that concluded that India came out on top, which is the closest thing I've seen to a thorough write-up by a Western source, and is probably be where I got my impression, at least in part. (Interestingly, as I recall, the Indian chest-thumping accounts seem to focus on the later Indian actions, which suggests strongly to me that Pakistan treated them pretty roughly at first and would corroborate CHPM's account.)
However, I'm not sure if CHPM's account is actually all that great – it's been a while since I looked at it (sorry!) but I think I recall feeling like it was not the final word on the matter – maybe it relied too much on biased sources or guessing for my taste, I'm not sure.
If you've got a good write-up or source, I'd be very interested in reading it.
Shrike
5d ago
This alongside the parallel discourse, also loud and proud on TheMotte, of men saying that in all justice the life of a woman should be to bear as many children as possible in quick succession, then look after them 24/7, while being the material provider (read: office-job attendance, open evenings, intact body, full nights of sleep) is the maximum expected from a husband.
I just want to flag a few things:
- If your wife is having kids nonstop, you aren't going to be having the full nights of sleep unless you are a literally absent father (as in, in a different house) and you aren't going to be having open evenings either (at least not all of the time) unless you are a figuratively absent father (as in, tuning out after dinner and refusing to help with the kids at all).
- Guys are more likely to have their intact body become not that way on the job compared to women – however the rates are pretty low.
- Notwithstanding the above, taking care of kids nonstop is hard work and while I suspect women are at least marginally better suited to it than men (and men likewise at least marginally better suited to e.g. office jobs than women) I think it's at least in some ways "harder work" than a 9 - 5, at least when you are dealing with multiple children, of which at least one an infant or toddler. Having "as many children as possible in quick succession" means that this is pretty much always the case.
I don't think the Motte's trad/pronatal corner is (generally) calling for exactly what you are suggesting (in part because many of them actually have kids and spouses that they actually like) and are aware both that women should not be treated as child production facilities and that in functional households the man bears part of the burden of parenting, just the same as the woman. (But if you are reading Terracotta's post and thinking "heck yes, that's exactly how things should be, you forgot the gigachad image" then, well, I think you are wrong.)
Shrike
5d ago
Is there any good understanding of how the recent dust-up with Pakistan went?
The sort of muddy impression that I have right now is that Pakistan gave India a bloody nose and India turned around and drop-kicked Pakistan across the ring. But finding objective Western documentation of the conflict seems difficult.
Shrike
5d ago
The Russia-Ukraine war demonstrates that second world armies facing second world air defences don't fly fixed-wing warplanes in anger, so destroying them doesn't hurt Iran much
What? There have been lots of fixed-wing warplanes operating on both sides in the Russia-Ukraine War on both air-to-ground and air-to-air missions.
Shrike
5d ago
If they all hit, then the war will be over in four hours. (But note that Raytheon says they intend to scale up to 1,000/year.)
It's hard to tell how much is enough because a lot of it is about the timeframe (1/year is fine if you started building around the Jurassic period, for instance), effectiveness, intended target, delivery vehicle...much of which isn't know.
My thinking is that China's window is closing, but they have a respite with the Iran war. Right now it looks to me like that window is about five years, but that's a guesstimate and could change due to further actions by the United States or China or third parties.
I think this is, at best, an oversimplifcation. If technology increases individual power per individual symmetrically, then the relative power of the individual to society and to other individuals will not change.
But this is not what we observe. At a minimum, technology benefits individual asymmetrically based on their skills. And what we in fact observe is that some technologies really do not increase individual power (particularly relative to others) since they require coordinated action to utilize. Thus some technologies are centralizing and shift the balance of power away from the individual and towards coordinated action, while some technologies are decentralizing and shift the relative balance of power towards the individual. However, even in the cases of decentralizing technologies, the societal collective generally retains more power than the individual by virtue of having more individuals.
The problem with things like "AI-designed viruses" isn't a problem of power, exactly - I would frame it more as a question of the relative strengths of offense and defense. (To explain a bit while I would say that it is not exactly a question of power, a person who has designed a virus still lacks a great deal of power that society retains - perhaps he has the power to kill everyone on Earth, but he still cannot construct an aircraft carrier, for instance. Whereas presumably society could construct both the carrier and the virus, giving it more power as long as it exists).
Some technologies (at least in the military sense) favor the defensive, while some favor the offensive. However, all else being equal, the offense always has an edge over the defense simply because the offensive has greater initiative. Thus the problem with things like "AI viruses" is that the technology is both individually empowering and is feared to favor the attacker.
However, technology will not inevitably favor the attacker. For instance, if you will forgive a toy explanation, think of missile guidance technology. Initially, this favored the attackers. Over time, though, the curve of that technology began to tilt towards the defender, because the technology in its early stages was very good at offensive tasks (hitting static or large targets) and bad at defensive tasks (hitting small, moving targets). But as the technology matured to hit small, moving targets, missiles began to be suitable for defensive tasks. If we imagine a future where missiles are perfectly good at hitting both large, static targets and small, moving targets, then it will favor the defender, since for an equal or lessor expenditure in defensive missiles he can perfectly hit all attacking missiles. Thus, while the attacker will always have the benefit of initiative, technological advancement can favor the defender.
It is not clear to me that the future of technology will favor the attackers inevitably. Many people feel personally empowered by AI, and thus see it as an increase in individual, offensive power. But the technology itself (at the cutting edge) favors coordinated action, resource allocation, funding, and control. Thus it seems possible to me that the long curve of AI technology may end up radically favoring centralized defenders over individual attackers.
More options
Context Copy link