Hacker News Reader

Comment by dooglius | Hacker News Reader

> unforgeable key internal to their hardware, like we have with PGP / GPG

PGP involves a private key, and if you have the private key you can "forge" any message. If you put the key in hardware, it can be read by an adversary with access to a powerful microscope.

View on HN

I mean how is this any different than the "secure enclave" on iphones or other forms of hardware security modules. Yes a sufficiently advanced adversary with an electron microscope could possibly extract the key but it still greatly raises the bar for 99% of other would be abusers.

Because the sort of actors who would try to forge videos are precisely the sort of actors who would have such advanced technology. The secure enclave on an iphone protects against say someone trying to convert a stolen phone into a stolen identity, not against nation-states.

what if the camera is connected to the internet, generates its own random key every 10 seconds, signs the new public key with the previous key, and a quorum of receiver citizens selected by sortition send their public keys, and the camera uses threshold cryptography to send each receiver their share of the secret frames for ~10 seconds.

The adversary would have to extract the key within 10 seconds without damaging the security envellope of the device, which can't be powered down. If such a camera is powered down, a replacement camera would need to be manufactured and sent and isntalled (again by citizens selected through sortition) at the place the malfunctioning / perturbed camera was. If the cameras cover each other (say cameras along both sides of a street such that a camera sees 2 or more other cameras) the perturber can be tracked, both where he came from, as well as where he went to...

But each device could have it's own unique signing key, signed by a manufacturer master key that is not distributed. Then you could revoke just that one device if the need arises (or even a manufacturer). Isn't this a fairly solved problem in the TLS/CA space?

How would you know when to revoke a key?

You seem to know more than you are admitting via your responses here. Perhaps you can share your thoughts, rather than provide responses which defeat the discussion?

It's a genuine question, I don't have an argument for impossibility or anything, I just don't see how revocation could practically be done. In scenario A, Joe Schmo buys a camera, happens upon a powerful politician in a compromising situation, and uploads the video to YouTube with a digital signature. In scenario B, Joe Schmo buys a camera, actually works for the NSA, uses an electron microscope to extract the private key, and stamps a digital signature on a forged video, uploading that instead. How would any third party be able to differentiate between those? I could imagine potential avenues for solutions (e.g. maybe you could use quantum entanglement to make some sort of tamper-proof chamber around the key in hardware?) but then we're pretty far afield of straightforward PGP/TLS. Not to mention the problem that an adversary like the NSA could just get inside the ASIC fab and copy keys from the machine that prints them.

make a distinction between cameras for fun, and cameras for undercover investigative journalism by journalists or citizen journalists.

the first type of camera is the one we have today, the second type would be more expensive, need to stay connected to the group consensus protocol, and need to stay powered, so journalists will be lugging extra batteries, and the camera would have 2 battery ports for switchover...

Your logic here could be applied to any form of cryptographic signing. How do we know when to revoke an SSL key? Someone could be misusing them without us knowing. But sometimes we do know.

When the key signs some embarrassing video it gets revoked as compromised. Simple as that

Reply on news.ycombinator.com