11M worth of chips is very small stakes for all involved[1]. It is just about two NVL72 racks worth.

Every logistics operations expects a certain percentage of theft, they insure for it. The cost of securing has to be less than cost of loss. The logistics departments are doing 10's of billions of dollars every month to every hyperscaler and all the neoclouds all over the country (and internationally).

Even military logistics routinely losses high value items - from the recent f-35 part they tried to ship via HK(and lost to China) to even full nuclear bombs that are lost to this day from the cold war era.

If we cannot secure nuclear bombs 100% during transit, there isn't going to be 100% success for computer chips.

[1]Even for foreign adversaries, probably not worth the risk of the scandal just to get two racks. They need orders of magnitude more - the South East Asian route in the news this week seems to be $5-10B worth of chips for context.

There's a lot more than just this incident, and not just GPUs.

https://www.tomshardware.com/tech-industry/big-tech/data-cen...

Maybe there's going to be a GTA 6 mission based on this eventually..

> probably harder to find a no-questions asked buyer

Is it ? with all the news this week about dummy servers and other tricks to bypass export controls, it seems there would be many no questions asked buyers abroad.

you don't need to, https://status.cloud.microsoft/. there is a public version

There are logged in pages for people to raise tickets, see specific services/regions they have access or restricted ones(Gov/Sovereign Cloud only available to some users) and also other features only logged in users would need.

Yeah it will not work under an auth downtime or the service application downtime, but it is not designed for that.

It is for for the items which are localized at service/DC/region level. The service app pages on azure covers planned maintenance, retirement of old services, security advisories and so on.

Why is Jupiter in the same category as planets then?

Jupiter does not orbit the sun (the first requirement per IAU definition of planet), the barycenter for the system is not inside the surface of Sol at 1.07 solar radius.

It is not planet then yes ? There is no push to promote it to a sub-brown dwarf or a new category .

Classifications are always arbitrary and only fits some bias one way or other.

A planet is an object that:

  a) orbits a star, and is not a star nor a brown dwarf itself;
     i) a star is an object that is massive enough and hot enough to conduct protium fusion in its core
    ii) a brown dwarf is an object that is massive enough to conduct deuterium fusion, but insufficiently massive (and hence insufficiently hot) for protium fusion  
  b) has attained hydrostatic equilibrium, that is, it is spheroidal;
  c) it has cleared its local neighbourhood of 'rival planets' since its birth;

Ergo, Jupiter is a planet. It is a certain type of planet—a gas giant—but a planet nonetheless.

There is some controversy about when giant planets stop being giant planets and start being brown dwarfs—typically around 30 Jupiter masses—but this is an active area of research, to understand the fusion ignition in these objects. As I mentioned in another comment, classifications are useful. Of course they are arbitrary and they are biased—they were decided by humans. But the entire point of science is to push the edges of our understanding and reduce these biases and make our models more general and less arbitrary.

Their claim is that it fails (a) because the Sun and Jupiter orbit a common barycenter, and hence Jupiter does not orbit the Sun (or any star).

Everything orbits a barycentre. The position of that barycentre within or without the primary is not really relevant within several orders of approximation—and when it does become relevant, then that system becomes interesting in itself. Chances are the planet in question is so massive relative to its primary that we need to ask a more fundamental question: are we even looking at a star-planet relationship, or is it closer to a binary star system? Then we can investigate how that system formed and developed, and generalise and improve our model of stellar/planetary system development.

There are plenty of binary star systems where this orders of magnitude standard is not met .

There is no academic literature or accepted consensus barycenter has to be orders of magnitude away - the accepted consensus is that a body orbiting another means internal barycenter, if they have external barycenter then they orbit each other period.

The fact that Charon and Pluto do not have internal barycenter is why Pluto-Charon (also Eris-Dysonmia) are considered dwarf planet systems ; and that was used partly in support (while not used in definition) of arguing for dwarf status of Pluto .

Yes — their point is that as-written, it orbits a barycenter external to the star: hence doesn’t orbit the star, but they’re in a mutual orbit.

Thus the suggestion for a new dwarf category — precisely because we do label times the barycenter is outside the star differently, as you yourself note.

You were so busy trying to “correct” me, you failed to address the topic.

'In orbit around' is frequently used as short-hand to refer to a secondary with mass << primary, to a first order approximation. In the Solar System, even for satellite transfers, the effect of Jupiter on the Sun can and is ignored.

The topic is pointless hair-splitting and nostalgia. It's like saying we can't ever use F = ma to describe free-body diagrams; we need a full relativistic per-particle QFT simulation to accurately and precisely model every single interation. That is impractical, at least for now. Likewise, we come up with classifications that are useful for the present time. If 'dwarf planet' is a catch all for all the small roundish bodies in our planetary system, then so be it. I suspect the IAU will make things more fine grained in the relatively near future. We even classify the asteroids as more than just rocky lumps.

We have QFT precisely because mechanics improperly categorized events.

We can similarly say that “cold dwarf binary systems behave similarly to planetary systems, due to the proximity of the barycenter to the star”.

And that’s more informative than calling Jupiter a planet.

I agree with this picture as the barycenter argument seals it for me. The only leap left is for me to believe Ol'Jove as a failed star (it still emits heat in excess of what it derives from Sol) and we find ourselves cozily herded about by a binary star system.

agents do the same thing - the recent Minecraft experiment was illuminating and anyone using enough of agent swarms/clusters have experienced similar things .

Maybe bureaucracy is emergent property of intelligence.

The CFOs he is talking about are the enterprises who are spending big money on the tokens ; they are not seeing RoI ; not hyperscalers or frontier labs

Yep, i can't disclose specific numbers but we went from $X/month/person budgets to $X-100/month recently, and this is a pretty big company with ~5000 people. From what I can tell the pursestrings are only tightening further too, too little actual real-world benefit for the obscene spend

% would be helpful, is it -1% or -100% per month per person?

Correct. The people that control the pot of money these labs ultimately need to survive seem increasingly unconvinced that giving these labs money is a good idea. At least at the levels the labs need to not go bankrupt.

Just a tad bit more ; it like saying either an event will happen or not so probability is 50%.

Your scenario is certainly possible, that doesn’t make both predictions equally (un)likely.

A similar incident would be the 1Billion robbery of the Bangladesh central bank

https://en.wikipedia.org/wiki/Bangladesh_Bank_robbery

Most of it was stopped and some was even recovered but $66Million is still lost 10years later .

In both incidents protection are reliant on the authorities to minimize the damage ; tether and circle able to freeze tokens is not that different from the traditional finance doing it.

Crypto is only reinventing structures that evolved in every financial system over time.

Overtime they are going away from their founding theme of decentralization . Exchanges and stable coin issuers who can freeze wallets and coins are by definition centralization ; it is worse though as these organizations are less transparent than banks with less accountability

And as the stable coin operators and exchanges grow closer with governments and political elites they will add more and more controls in exchange for legitimacy.

Azure (and AWS) offer Gov clouds, and probably on prem air-gapped deployment of their full stack of Fabric for that type of NEED security customers who can pay for it.

It is trade-off between risk and access/productivity with a cost dimension always.

Depending on your threat model you could go to typewriters /paper as some Russian department did famously few years back too.

I setup parts of govcloud.

The real sensitive customers are on oxide.

AWS (and others) have ladders of this was my point. GovCloud is just the one end of the this AWS also regions for Secret and Top Secret work which are reported to be air-gapped. There are probably programs beyond that where the customers are operating the stack fully on their own and so forth.