site banner

Transnational Thursday for October 31, 2024

Transnational Thursday is a thread for people to discuss international news, foreign policy or international relations history. Feel free as well to drop in with coverage of countries you’re interested in, talk about ongoing dynamics like the wars in Israel or Ukraine, or even just whatever you’re reading.

2
Jump in the discussion.

No email address required.

Somebody asked last time I posted a list like this what the background was. I'm coming from an EA/forecasting background, but then realized that although there might be something to being worried about catastrophic risks, reponses to this were top-down, trying to conceptualize risks long beforehand. I grew very unsatisfied with this, particularly for AI, and ended up raising some money to run a foresight/fast response team. We produce weekly minutes here, and the below feeds into that.

Some general topics:

  • Will NK detonate a nuclear weapon? When?
  • To what extent is ww3 a good level of analysis for global conflicts?
  • I used to not worry that much about climate change, but 100-1k people killed in my own backyard (Spain) makes me a bit more worried
  • On the one had, a terrorist cooking ricin is a bit alarming. On the other hand, it shows that Al Qaeda doesn't have the chops to do anthrax or bottox. Thoughts?
  • Is the WHO's global emergency corps bullshit? Seems like it's a "reserve of experts"
  • I didn't know that France depended on Rosatom for nuclear fuel. Lol.
  • We've been seeing mpox coming to developed nations for a while, but it's still striking to see the 1st london case.

South Korea’s military intelligence agency told lawmakers Wednesday that North Korea has likely completed preparations for its seventh nuclear test and is close to test-firing a long-range missile capable of reaching the United States.

An article looks at the growing alliance between China, Russia and other powers

Jamie Dimon, the head of the financial giant JP Morgan, makes the argument that we are already in a WW3.

Animal testing of H5N1 gives some data about how well it's adapting

Russia launches exercises simulating retaliatory strikes

Pakistan vows to emphasize military ties to Russia, and collaborates on anti-terrorism exercises.

At least 100 people have died so far (and about 1000 are "disappeared") in flash flooding of Spain’s Valencia. Bridges collapsing, and overall very striking videos on social media. The city got what would have been a freak tornado, but such events might become more common as climate change continues changing up weather patterns.

A teen who went into a murderous rampage was also cooking ricin.

Israel ordered a whole Lebanese city evacuated

Geneva convention rules are being weakened, and civilians aren't being shielded from the worst harms in Ukraine or Gaza.

Finland seized Russian assets over compensation linked to invasion of Crimea

A Boeing satellite exploded into 500 pieces. The worst case scenario in events like this is Kessler syndrome but so far reporting doesn't point to something like this, though early simulations don't look great

The US and China are fighting over dominance in the depths of the South China sea

The WHO activated the global emergency corps to deal with monkeypox. Implications unclear, as it seems more like a "reserve of experts that advise" and less like a "reserve of nurses and doctors"

A cyberattack from Iran hit an Israeli bank, and maybe credit card users generally, blocking users off.

Cyberattack against French Internet Service Provider

New agreement between Germany and the UK will tighten cooperation

Ballot box arson attacks in Oregon.

More cyberattacks in Australia

Fire in UK shipyard which builds nuclear submarines

The 2025 geomagnetic storm season might be pretty big

France depends on Russia for nuclear fuel

Some Russia military bases are empty. Some experts suggest this is for sabotage operations in the Baltics

First case of mpox Ib clade in London

Floods also caused havoc in Africa

Putin launches drills of Russia's nuclear forces simulating retaliatory strikes

India is expanding nuclear capabilities with fast breeder reactor

The US CDC issued an alert for "walking pneumonia"

A man with 120 guns and 250,000 rounds of ammunition in his home was arrested for shooting at a Democratic Party office

in Tempe

AI "Will Enhance" Nuclear Command and Control, Says nuclear command general

North Korea likely to ask for nuclear technology from Russia in exchange for troops, South Korea says. This would mirror an agreement between Iran and Russia.

The US army is preparing for a possible confrontation with China

Hezbollah new leader might agree on a ceasefire

More coral bleaching

Israel is using AI tools with little oversight to determine whether an individual is a Hamas operative.

North Korea conducted an ICBM test.

H5N1 detected in pig. Previously only in cows

A Boeing satellite exploded into 500 pieces. The worst case scenario in events like this is Kessler syndrome but so far reporting doesn't point to something like this, though early simulations don't look great

This is because it's in the wrong place.

Kessler syndrome is a threat in low Earth orbit - the region right around the globe in that video, up to 2000 km above the surface (note that this is an orbital radius of up to ~8400 km; you are closer to this region than you are to the centre of the Earth). It's a threat because we've put tons and tons of satellites there, on all sorts of different angles and thus at high relative velocities, so there's potential for debris to multiply over time (one satellite blows up, making debris that blows up 10 more satellites, which make debris that blow up 100 more satellites, etc.). Medium Earth orbit and high Earth orbit are far less cluttered, so there's not enough fuel for that sort of chain reaction. Geostationary Earth orbit (where this explosion happened) is a special case - we've put a ton of satellites there, but unlike in LEO, they're all going the same direction (prograde equatorial orbit), so the relative velocities are low and collisions AIUI shouldn't cause the snowballing effect that threatens LEO.

Another relevant fact is that the problem is self solving with time at any height where there is meaningful atmospheric drag because any debris would need further boosting to stay up.

Well, yes, although in most cases only after it destroys all satellites at that height and below. (There is also the requirement that people stop putting new stuff into the Kessler region until the junk does all deorbit - which one would hope would be the case, but who knows.)

Thanks, I had been wondering about this and this neatly explains it.

A few of your notes are missing links - was this an error?

On the one had, a terrorist cooking ricin is a bit alarming. On the other hand, it shows that Al Qaeda doesn't have the chops to do anthrax or bottox. Thoughts?

Ricin is more dangerous than anthrax, because anthrax responds well to treatment and ricin doesn't. There's this interview with Fauci, which I'm having trouble tracking down (I found another one where he talks about it, but it's not the one I'm thinking of), where he says that when he got an envelope that sprayed him with white powder, he immediately figured three possibilities, and his words were something like "1) it's baby powder, and I'll be fine, 2) it's anthrax, and I'm gonna get sick, but I'll take antibiotics and I'll recover, 3) it's ricin, and I'm dead".

Anthrax contamination can be expensive to clean up if there's a lot of it (infamously, the British tested it on an island, and cattle that ate grass there still died from anthrax 70 years later; the way they ended up fixing the site was to strip the topsoil off the entire island and incinerate it), but deaths are rare and epidemic's impossible (because modern Western countries don't leave dead bodies lying around and anthrax doesn't spread living to living).

I see, I'm more worried about incidents with many casualties, but it seems interesting that it's more lethal given infectoin.

To be clear, while generally regulated as a biological weapon due to its origin (it's extracted from the castor plant), ricin is a toxin rather than something that replicates; one is poisoned with ricin, not infected. Anthrax, however, is weaponised as viable spores, which do infect people and replicate inside them (but as noted, anthrax doesn't spread living to living; it's corpses of people/animals that have died of anthrax that spread it to others).

Bioweapons that actually threaten epidemic are generally known to be For Crazy People Only (because, well, once you release one of those it's pot luck whether it comes back to infect your guys too). Stuff like plague (ISIL played around with this a few years back, IIRC, but thankfully they only killed themselves with it) and smallpox (extinct, but the tech exists to bring it back) fall into this category.

Thanks!

Israel is using AI tools with little oversight to determine whether an individual is a Hamas operative.

I'm not sure what is actually being reported here. So far I see two facts being alleged - that Israel is using AI system to figure out who could be Hamas operative, which is 90% accurate (spectacular number if true, to the point I even suspect they are being over-optimistic), and that Israel is using phone tracking to locate specific suspects. The latter has nothing to do with AI, as for the former - I am not sure what is supposed to happen after a certain person has been identified as "90% likely to be Hamas operative". The article uses phrases like "automated kill chain", but there's no evidence or even allegation such thing actually exists in any meaning of the word "automated" - do they mean if the system identifies a person, he would be automatically targeted by some killing machine without human supervision? If so, why don't they say it explicitly and describe what this system is and how they know about it? If no, then what "automated" means?

And on the other hand, I am not sure what kind of oversight you would put on such a system. Let's assume you indeed had a system which with probability of 90% can tell you whether or not certain guy in Gaza is in Hamas. Now, how would you verify it? Obviously, if you had some better system, you'd use that one from the start. You could review the data yourself - but do you have better than 90% accuracy? I mean, if you spot some hilarious bug in the system - on the level of "black vikings" and other hilarious bugs in public LLMs, sure. You can block that. Like if the system marked every guy with name "Muhamad" as Hamas member, than you can notice it and overrule the system. But let's say you didn't notice that. Moreover, you tried it 100 times and went out and captured those guys and 90 of them admitted that yes they are in Hamas, or you found Hamas membership card on them and so on. E.g. let's assume 90% is true. How do you oversight that system then? Verifying each person manually is impossible - there are like 50 thousands of them, and most of them are hiding and it's impossible to verify anything about them until they are either captured or dead. So what do you base your supervision on?

A cyberattack from Iran hit an Israeli bank, and maybe credit card users generally, blocking users off.

I couldn't get from that - which bank was that and what actually happened?

The first one has no text and is an assembly of half-second clips with rapid cuts, can you make any sense out of it? Like, figure out what is the bank's name and what happened to it?

The second one at least gives the name of the company Automated Bank Services, and the source of their reporting: https://www.timesofisrael.com/liveblog_entry/credit-card-transaction-company-says-mornings-payment-problems-caused-by-attack/

So yes, it looks like they could indeed mess up with Israel's credit card processing provider (not a bank), blocking it for about 3 hours. Looks like a banal DOS, for which they had no protection for some reason.

I appreciate these sort of roundups and what you're doing. Cheers.

Thanks! I appreciate The Motte reading my shit and running with it

The undersea SCS sensor stuff is interesting, make a lot of sense for both sides to lay sensors there. I wonder if it would make sense to just fill the whole area with them willy-nilly, dumping them out of ships and aircraft just before going in? For China, if both sides can see eachother's submarines and their missile subs are safe in the Sea of Bohai, it might even the gap.

The Spain floods have nothing to do with climate change. They’re a once-every-fifty-years occurrence on that floodplain and have been for all of recorded history, at least for centuries. The last major flood occurred in 1957, after which Valencia (the city) got flood defenses. It was spared the worst this time, which happened in the environs and surrounding smaller settlements.

They’re a once-every-fifty-years occurrence on that floodplain and have been for all of recorded history, at least for centuries.

Because the rain in Spain stays mainly in the plain?

Sorry, I couldn't resist.

Out of all these things, I'm most worried about H5N1 over the short and mid term.

If it transmits between pigs (soon will, I assume) rather than just infecting one here and there, I'm led to believe the road to a human to human transmitting virus pandemic is a short one...

I think I'm mostly focused on terrible outliers, and the mortality for H5N1 just doesn't seem that high. Say 0.05x to 3x covid, something in that range. You then have to add long-term effects, which are hard to estimate &c, but it just doesn't seem very existentially threatening.

Depends on the variant and how it mutates. If it becomes an easily transmissible virus that attacks the lungs it could get much worse than covid. Ferrets, who are quite similar to humans as far as respiratory system goes, afaik, have a terrible death rate to h5n1.

Yeah I think the medical side of things are most worried about the flu side of things. Monkeypox doesn't excite me, Myco isn't a big deal.

A Kessler cascade is one of my biggest fears though, yikes.

Why doesn't monkeypox excite you? It could fill in the niche that smallpox previously had.

Yeah. That's one of my fears too. The more trash we put into orbit, the higher the risk goes of making the satellite space unviable forever.

Boeing keeps winning.

I think a nuclear war between the United States/NATO and Russia is still the most likely existential crisis facing this planet. It doesn’t get nearly as much as attention as it should, here or elsewhere, because it’s now old fashioned and unstylish. And because younger people haven’t read about the likely scenarios for how it would happen, so the don’t realize the immense time pressure involved. The leadership on both sides would have about a ten minute window to make the decision whether to torch the planet. I find it worrying how cavalier everyone is about this, all the way from Reddit to the high political leadership.

Why you expect it to be a high risk? Even if Russia would nuke Berlin and Warsaw - then USA, France and UK would not nuke Moscow back.

And reaction would be likely strong enough (China would join) ensuring that Russia will not nuke NATO countries or even Ukraine, as benefits is not there.

And if Putin/Biden/Kamala/Trump would be insane enough to launch full scale attack... There is not much we can do with that.

It doesn’t get nearly as much as attention as it should, here or elsewhere

Maybe because when I seen it, it was in "Russia has nukes, we must satisfy all their demands" form which does not encourage treating it seriously.

existential crisis

Even full scale use of all nuclear devices would not cause end of civilisation: it would cripple it and set us 50-100-200, maybe 300 years back and kill billions - but would not end existence of humanity.

Yes, people claiming that we have enough nukes to kill all humanity were stupid and/or lying.

Maybe because when I seen it, it was in "Russia has nukes, we must satisfy all their demands" form

If you believe the recent Woodward's book, that has been the essence of the US policy towards Russia for the current administration. They were all overcome by mortal fear of Russia using the nuke, likely due to bad intel (probably injected by Russia). And since Woodward is pretty much a sock puppet for the people who define that policy, I think it is believable.

But if we were set back to the technological level of the 1700s, how possible would it be for us to recover to our modern-day level? Most of the easily-accessible coal and oil have been depleted. Modern farming and transportation would be destroyed, very possibly leading to Malthusian living conditions and a lack of leisure time necessary to rebuild the machines and infrastructure that we know are possible. Within a few generations, most of the practical knowledge necessary to build complex machinery, etc., would be lost completely. It also seems likely to me that in the aftermath of a large-scale nuclear war, many of the most intelligent people would be dead, which might further complicate efforts to build society back to its present level of affluence and technological advancement.

Most of the easily-accessible coal and oil have been depleted.

offset by known locations of some remaining, some in progress mining operations being left and technology being invented already

Modern farming and transportation would be destroyed

not all of it, enough to bounce back

very possibly leading to Malthusian living conditions and a lack of leisure time necessary to rebuild the machines and infrastructure that we know are possible

seems dubious, but it is unverifiable and there is no known way to verify it

Within a few generations, most of the practical knowledge necessary to build complex machinery, etc., would be lost completely.

seems extremely dubious, I would buy utterly losing CPU production... But outright losing most machinery production? It is not SO hard to bootstrap when you are motivated, knowledge exists and world heavily fragmented. At least somewhere you will have people going to rebuild. And that is even if someone went omnicidal stupid and nuked say Peru and Keyna instead of finishing their main rival. Just to spite survivors.

It also seems likely to me that in the aftermath of a large-scale nuclear war, many of the most intelligent people would be dead

many of [insert group] would be dead, for nearly all descriptions

but even with selective death of top 40% of most intelligent/hard working/insert descriptor here survivors would rebuild. I think that you underestimate resilience of humanity (or maybe I overestimate)

But if we were set back to the technological level of the 1700s, how possible would it be for us to recover to our modern-day level?

Very. Null risk.

I've cited Ord previously on this topic, but I'm feeling lazy today so I'll just quote:

Even if civilization did collapse, it is likely that it could be reestablished. As we have seen, civilization has already been independently established at least seven times by isolated peoples.12 While one might think resource depletion could make this harder, it is more likely that it has become substantially easier. Most disasters short of human extinction would leave our domesticated animals and plants, as well as copious material resources in the ruins of our cities—it is much easier to re-forge iron from old railings than to smelt it from ore. Even expendable resources such as coal would be much easier to access, via abandoned reserves and mines, than they ever were in the eighteenth century. 13 Moreover, evidence that civilization is possible, and the tools and knowledge to help rebuild, would be scattered across the world.

13 Overall the trend is toward resources becoming harder to access, since we access the easy ones first. This is true for untouched resources in the ground. But this leads people to neglect the vast amount of resources that are already in the process of being extracted, that are being stored, and that are in the ruins of civilization. For example, there is a single open-cut coal mine in Wyoming that produces 100 million tons of coal each year and has 1.7 billion tons left (Peabody Energy, 2018). At the time of writing, coal power plants in the US hold 100 million tons of ready-to-use coal in reserve (EIA, 2019). There are about 2 billion barrels of oil in strategic reserves (IEA, 2018, p. 19), and our global civilization contains about 2,000 kg of iron in use per person (Sverdrup & Olafsdottir, 2019).

This is, of course, leaving aside the issue that substantial chunks of the world would be directly untouched by nuclear war (Africa/South America, also probably New Zealand and Ireland), so it's not exactly like literacy will be lost forever in 20 years or something even if rebuilding fails in all the places that are involved.