Hacker News Reader

Comment by bonchicbongenre | Hacker News Reader

The answer: yes, they could've. I had a professor who previously had been part of a US gov group that had been tasked with preparing for an attack on the electrical grid, all the way back in the tail end of the cold war. His impression of the security of the US grid was that it was completely unsecured. He told me that they had no chance of solving the problem then, and that he expects the same is still the case. His worst fear is a large EMP attack, not locally cutting powerlines, but the danger still stands

View on HN

Putting aside EMP attacks (a real risk, but harder to mitigate against), the big issue right now on a day to day level is that energy suppliers are under pressure from regulators to reduce prices to consumers (in most countries which run regulated private sector energy distribution).

This means redundancy is written down by MBAs as excess capacity. It means assets are stretched to far nearer their safe limits than was ever envisaged previously. This ironically reduces their safe working life (in many cases), but this problem is cast along the road - in markets where providers re-tender regularly to continue to operate, that means it's a problem for the next franchise holder.

The incentives just aren't there to encourage creation of new redundancy. Existing redundancy won't be removed (as that would cost more!), but the long term outcome of this kind of "regulated private provider" system seems to be systemic under-investment in capital assets, with a view to deferring the problem until another time.

Another big challenge is that often the regulators for energy companies are "market" regulators, and therefore staffed by expert economists, rather than expert engineers. This further perpetuates the myth you can maintain and run critical national infrastructure with an economics degree and a calculator, and means the regulator tends to focus solely on economics, while assuming any issues with technology are fixed through economic incentives. Often they aren't, as the cost of a systemic black swan event is multiplied out by its likelihood, and the end result is to take no action while enjoying the profits today.

At least the above is based on my experience in trying to get energy market regulators to understand security and the issues of their regulator approach.

That’s weird that his concern is an EMP attack. Pretty much every serious nuclear scholar I’ve read find an EMP attack laughable. It’s a great thing to stir up fear among a domestic audience, but as an actual military tactic, it’s stupid. First, there’s never been an operational test anywhere, second, you have to fire a nuclear weapon, which is the very act that triggers a full scale nuclear response because of launch-on-warning. And finally, there’s just better uses for a nuclear weapon, like you know, air bursting in a city.

An EMP attack from an enemy nation is unlikely but a solar flare could happen at any time. In 1859 the world experienced an enormous geomagnetic storm when a coronal mass ejection hit Earth's magnetosphere. If the same solar flare happened today most of the world's power grid could be wiped out minus government and military hardened assets.

Non nuclear EMP weapons have existed for years, and that's only the stuff they're talking about publicly. There may be other techniques, say distortion of the Earth's magnetic field to send charged particles toward a target, that we don't know about.

The non-nuclear EMP weapons are very weak and have only minor localized effects. They aren't capable of taking out the power grid for a whole city or something.

And please no one waste our time by claiming that the military has some secret magic EMP technology. The laws of physics impose severe constraints on what can be achieved with conventional explosives in a weapon small enough to put in an airplane or truck.

You only need to take out a few key points to take out the grid. Large scale EMPs are not needed. Knowing which key points to take out is the domain of nation states, but it is absolutely possible.

> there’s just better uses for a nuclear weapon, like you know, air bursting in a city.

If your choice is completely annihilating one city, or destroying even just 10% of computers (including industrial control systems) in 50% of the country, which do you think causes more damage?

Destroying even a small fraction of individual computers makes most larger systems inoperable.

Without industrial control systems, power plants can't operate. Without power, cities become unsurvivable for a majority of the population within days, industrial output becomes effectively zero, and immediate issues like starvation and resulting unrest become a much bigger priority than fighting a war. Most importantly, recovery efforts will be extremely slow without power. How do you call the supplier, or other places that might have spares, when there is no phone network, no cell phones, no Internet? How do you call the experts that could diagnose which of the hundreds or thousands of control components need swapping? How do they get the fuel for their vehicle to get to you? And food and water to survive until you put everything into place? And safety so they don't get murdered by the looters?

According to http://www.empcommission.org/docs/A2473-EMP_Commission-7MB.p..., a peak field strength of 15 kV/m is expected to cause permanent damage to computer equipment. https://en.wikipedia.org/wiki/Nuclear_electromagnetic_pulse#... shows a single explosion (of unspecified yield) covering most of the US with a 25 kV/m field.

> which is the very act that triggers a full scale nuclear response because of launch-on-warning.

AFAIK the US doesn't practice launch-on-warning. Even if it did, a peaceful low earth orbit satellite launch is hard to distinguish from a not-so-peaceful low earth orbit nuke launch (unlike a conventional ICBM that's supposed to come back down). It would likely still result in nuclear retaliation, but if e.g. a dictator was already being invaded by the US, they don't necessarily have much to lose.

You are very misinformed about how satellite launches are publicized, and about the US launch on warning policy. Even if you are too young to remember the Cold War, a simple google search will confirm this.

Again, the threat of an EMP attack is absurd, perpetuated by fundamentally not serious people.

https://foreignpolicy.com/2020/07/21/electromagnetic-pulses-...

> You are very misinformed about how satellite launches are publicized

Then tell me, where is the difference between North Korea saying "we will be launching this peaceful satellite of peace" and launching a spy satellite, and North Korea saying the same and launching a nuke satellite? Same orbit, same weight, contents kept secret to the best of their ability (and possibly swapped last minute in secret), same orbital timer to trigger it, only difference is whether it takes pictures or goes boom.

> and about the US launch on warning policy

https://www.armscontrol.org/act/1997-11/news/clinton-issues-... states:

Bell pointed out that while the United States has always had the "technical capability" to implement a policy of launch on warning, it has chosen not to do so. "Our policy is to confirm that we are under nuclear attack with actual detonations before retaliating," he said.

> https://foreignpolicy.com/2020/07/21/electromagnetic-pulses-....

This just says that a nuclear EMP attack would start nuclear war (or rather, nuclear retaliation if the EMP nuke was the only one the attacking country had/could deliver).

An EMP attack by Russia is indeed unlikely.

Reply on news.ycombinator.com