On the topic of Faraday cages: is there a way to make one whose "Faradayness" can be turned on and off at will? I remember hearing that you can do so with current, but I haven't found much info since.
For my money, a triple burrito wrap of foil and ziploc gives 90+ dB attenuation plus moisture and mechanical protection.
Also nitpick, but the E1 pulse doesn't fry beefy transformers, instead it's the long-period E3 pulse (seconds to minutes) that raises the grid voltage and overvolts the transformers.
I did a much lower fidelity experiment this past week with one of my kids.
I have some radios I threw in metal boxes on the off chance something like this happened so we’d have some way to get emergency alerts. To check how well it shielded we threw my phone in it and they called to see if it rang. It did.
So we tried some other things - the microwave (worked), a tool chest (didn’t), a faraday bag (worked), miscellaneous storage boxes (didn’t).
Fortunately, the box with the radio was an old box and some paint had worn from the edges. We made a “seal” out of aluminum foil and tried again. It worked.
This only tested high frequency signals and probably doesn’t reflect real EMP risk in any way, but it was fun to do.
When you say "worked", do you mean the phone worked in the microwave or that the microwave worked as a Faraday cage and prevented the phone from ringing?
Technically I'm measuring in the "wrong direction" since I would be very surprised if an EMP were initiated inside a trash can in my basement, but shielding works the same in both directions so this doesn't matter.
It really doesn't, with 1.5GHz you're still in near field regime inside the can.
When apple started doing mac address randomization on the iphone, we used a metal trash can to test that in a controlled environment. It worked quite well for wifi isolation. We did not tape the lid.
Since you already suspect that any EMP will come from above why didn't this guy simply turn the trash can upside down and place it over the test setup?
There would be no need to seal anything since the can has a continuous rolled seam. Just use a standard trash can instead of one of those kitchen trash cans that likely has a plastic bottom for easy clean-up.
I use these as planters for my orchard fruit and nut trees. They have no open holes. I have to drill the bottoms or side to add drainage so that the trees don't drown.
This is a cool test and a great use of (relatively cheap) rtlsdr dongles. Ignoring the "prepper" aspect a faraday cage is genuinely useful thing to have around if you want to play with radio transmitting without any chance of breaking any laws.
A much better faraday cage is matroiska doll nesting of two different size trash cans with a non-conductive platform supporting inside smaller can. The air gap and double conducting surfaces really help. And with the different circumference of the smaller can's rim there'll be a different resonant freq that gets let in. So while the outside seam may pass ~339 MHz without any attenuation the inner seam might pass ~450 MHz and so both unintentional passbands will get at least the minimum ~20dB attenuation per can.
This kind of prepping is silly. I suppose it's a way for adults to build pillow forts.
AI isn't likely to deliberately kill anyone. It's likely to make them entirely economically uncompetitive.
Today, if people go on strike, the economy grinds to a halt. This gives them a good deal of political power. If the economically useless go on strike, nobody cares.
That leaves violence to make people's voices heard, and the surveillance and police forces that we've been automating should be effective at undermining, dispersal, or suppression.
Automating the economy is the risk to humans, not EMPs.
I agree with you that this is silly, and don't recommend it. It looks to me like even in an EMP situation, you don't need a Faraday cage, you just need your equipment to be unplugged: "Overall, I think it's pretty likely that for EMP resistance it's enough to leave your equipment off and unplugged. Unless you're excited about testing Faraday cages, make sure your stuff stays unplugged and you can stop reading now."
On the other hand, I don't think we can rule out a lot of situations in which an EMP might be used, even if none of them involve an AI directly trying to kill people. For example, other countries could conclude that risking nuclear war is a better bet than allowing the US to build superintelligent AI.
On the topic of Faraday cages: is there a way to make one whose "Faradayness" can be turned on and off at will? I remember hearing that you can do so with current, but I haven't found much info since.
Nice to see the results match Dr Arthur Bradley, the OG in the field (no pun intended).
https://www.youtube.com/watch?v=aMct99DiZak
For my money, a triple burrito wrap of foil and ziploc gives 90+ dB attenuation plus moisture and mechanical protection.
Also nitpick, but the E1 pulse doesn't fry beefy transformers, instead it's the long-period E3 pulse (seconds to minutes) that raises the grid voltage and overvolts the transformers.
I did a much lower fidelity experiment this past week with one of my kids.
I have some radios I threw in metal boxes on the off chance something like this happened so we’d have some way to get emergency alerts. To check how well it shielded we threw my phone in it and they called to see if it rang. It did.
So we tried some other things - the microwave (worked), a tool chest (didn’t), a faraday bag (worked), miscellaneous storage boxes (didn’t).
Fortunately, the box with the radio was an old box and some paint had worn from the edges. We made a “seal” out of aluminum foil and tried again. It worked.
This only tested high frequency signals and probably doesn’t reflect real EMP risk in any way, but it was fun to do.
When you say "worked", do you mean the phone worked in the microwave or that the microwave worked as a Faraday cage and prevented the phone from ringing?
They are saying the microwave and the faraday bag worked as a faraday cage.
The storage boxes and tool boxes did not function as a faraday cage.
Technically I'm measuring in the "wrong direction" since I would be very surprised if an EMP were initiated inside a trash can in my basement, but shielding works the same in both directions so this doesn't matter.
It really doesn't, with 1.5GHz you're still in near field regime inside the can.
He’s testing penetration from the opposite side regardless of the source location.
Yeah, you put a transmitter inside the can, and a receiver inside the can, and the receiver is gonna get signal.
When apple started doing mac address randomization on the iphone, we used a metal trash can to test that in a controlled environment. It worked quite well for wifi isolation. We did not tape the lid.
Since you already suspect that any EMP will come from above why didn't this guy simply turn the trash can upside down and place it over the test setup?
There would be no need to seal anything since the can has a continuous rolled seam. Just use a standard trash can instead of one of those kitchen trash cans that likely has a plastic bottom for easy clean-up.
That wouldn’t help I don’t think. I’ve had very small holes in a faraday cage results in rf leaks - slots thinner than a credit card.
I see what you are saying and understand that problem but I was picturing something like this type of can from Home Depot:
https://www.homedepot.com/p/Behrens-31-Gal-Galvanized-Steel-...
I use these as planters for my orchard fruit and nut trees. They have no open holes. I have to drill the bottoms or side to add drainage so that the trees don't drown.
This is a cool test and a great use of (relatively cheap) rtlsdr dongles. Ignoring the "prepper" aspect a faraday cage is genuinely useful thing to have around if you want to play with radio transmitting without any chance of breaking any laws.
A much better faraday cage is matroiska doll nesting of two different size trash cans with a non-conductive platform supporting inside smaller can. The air gap and double conducting surfaces really help. And with the different circumference of the smaller can's rim there'll be a different resonant freq that gets let in. So while the outside seam may pass ~339 MHz without any attenuation the inner seam might pass ~450 MHz and so both unintentional passbands will get at least the minimum ~20dB attenuation per can.
This kind of prepping is silly. I suppose it's a way for adults to build pillow forts.
AI isn't likely to deliberately kill anyone. It's likely to make them entirely economically uncompetitive.
Today, if people go on strike, the economy grinds to a halt. This gives them a good deal of political power. If the economically useless go on strike, nobody cares.
That leaves violence to make people's voices heard, and the surveillance and police forces that we've been automating should be effective at undermining, dispersal, or suppression.
Automating the economy is the risk to humans, not EMPs.
[author]
I agree with you that this is silly, and don't recommend it. It looks to me like even in an EMP situation, you don't need a Faraday cage, you just need your equipment to be unplugged: "Overall, I think it's pretty likely that for EMP resistance it's enough to leave your equipment off and unplugged. Unless you're excited about testing Faraday cages, make sure your stuff stays unplugged and you can stop reading now."
On the other hand, I don't think we can rule out a lot of situations in which an EMP might be used, even if none of them involve an AI directly trying to kill people. For example, other countries could conclude that risking nuclear war is a better bet than allowing the US to build superintelligent AI.
Is it also a risk to the humans automating it?
If AI works out as the frontier labs hope, humans will not be involved in automating it -- the improvement will also be automated.
So, of course not. Those humans won't exist.
Seems easier just to kill the humans, and then not bother with that complicated multi-step process
Why would any AI care about killing people? It's being engineered to make companies money.
it's not "engineered", it's grown via GSD in a process no one understands well