SN 920: An End-to-End Encryption Proposal

Beep boop - this is a robot. A new show has been posted to TWiT…

What are your thoughts about today’s show? We’d love to hear from you!

Regarding the misuse of hacking tools, Steve said what possible legitimate use could they have? There are a lot of hacking tools out there that have legitimate purposes, they are designed to be used by so-called “Red Teams” to break into company networks, at the network owner’s behest

They use a lot of tools that fall under hacking tools and their use is allowed in many countries, as long as they are only used for legitimate cyber security work - the company doing the hacking has to have a watertight contract with the company they are “hacking”, otherwise it might cause problems. Steve has reported on this several times, so I was a little surprised.

Regarding the routers with credentials story, we remove all storage media, before we dispose of our hardware. If we repurpose them, for example the wife of one of our CEOs (we have 3) works at the local library and old PCs are often donated to the library or local schools, the old hard drive or SSD is always removed and securely disposed of and a new SSD is placed in the PC and Windows re-installed, before it is handed over.

At a previous employer, they were a factory manufacturing company - they built new liquid chemical factories for customers - we would take the old hard drives out to the workshop and use the stand drill to drill multiple holes through the platters, for SSDs, they were physically smashed. At my current employer, we have a certified disposal company take them in bulk and destroy them - we can observer, and we get a destruction certificate after the fact, which is required as part of some of our industry certification programmes.

Routers etc. are reset and then securely disposed of in the same way. If the devices are dead and can’t be reset, we usually physically remove the storage card, if possible, or if it is chips soldered onto the motherboard, the chips are physically destroyed in place.

1 Like

Another thing, Steve asked if ChatGPT stores conversations… That was how the Samsung data was found. Likewise there have been cases of managers putting their subordinates’ personal data into ChatGPT & asking for a summary, only for the information to appear for all to see.

Likewise, a CFO entered confidential financial data into ChatGPT and asked it to produce a slide deck, ChatGPT then leaked the data.

Even Google got involved a couple of months back, banning employees from using it, because their proprietary code was starting to pop up in results from ChatGPT or one of its offspring.


ChatGPT now offers the ability to not use the information you enter into it to train it further, if you turn off history. In the default state, the information you enter will be used to train the AI further, which means it “knows” what you have entered into it, when it is talking to others.

1 Like

I think Steve’s proposal that subverted encryption be only required on minors is a good one. Thank you for being willing to share it.

It won’t prevent all possible problems, but no practical solution solves all possible problems.

I think it would have a powerful positive effect. First of all, it would help counter grooming, and more importantly, it would help counter sending of porn to and from minors. I took training from the Boy Scouts of America (BSA) on how to look out for these predators. Many predators will try to send minors porn as a way to desensitize them, and will try to convince the minors to send sexual pictures. So such a law would directly counter the predation and initial steps. More importantly, it would act as a strong deterrence for creating & sharing new child porn that comes from a child.

Sure, this doesn’t directly prevent adult predators from sharing child porn amongst themselves, but passing a law isn’t going to do a lot either. Predators can simply use an encryption system not legally available in their country, after all, they’re already doing something illegal. The primary impact of the currently proposed laws won’t be to catch child porn; their primary impact will instead be to eliminate privacy of adults.

If the lawmakers actually want to protect children, instead of grandstanding, let’s focus on the children. Once you’ve reached the age of majority, you should be able to have privacy.

1 Like

The other way to take this line of thinking is: treat all children as criminals, and monitor them likewise.

The problem the government wants to address has nothing to do with the children beyond them being victims. The crime they want to suppress is ADULTS trading it very ugly pictures of children where, most often, the children were forced/manipulated into making the content with adults. I don’t have any solutions simply because I don’t believe any can exist. There is NO way to police this and still have privacy, and the government’s real goal is that you have no privacy.

Where would the end-to-end encryption breaking, if it occurred, actually happen? Would the governments force themselves into the head-ends of the services they want to monitor? Force the ISPs to monitor? If the latter take the Firstnet public safety mobile network for example, that network is touted as being end-to-end encrypted at the network level…

Most likely they’d do something like: We have defined an agency, let’s call it the Office of Government Monitoring of Public Messaging (OGMPM). Here is the OGMPM’s public key. For every message you send, you must format secondary messages (known as the “monitoring messages”.) The first monitoring message contains metadata about the user’s message: The Sender Name, IP Address, Time, Recipient Name, IP Address, Time, and the end-to-end (symmetric) encryption key used to encrypt the message. You then encrypt the monitoring message with a random symmetric key using AES256, you then create the second monitoring message when you encrypt the symmetric key of the first monitoring message with the OGMPM public key. You then send these two monitoring messages to the OGMPM, in real time, via the following means (which I wouldn’t presume to define or explain here.) You MUST ensure the auxiliary messages is confirmed delivered to the OGMPM BEFORE you deliver the user’s message to its recipient.

This is how they are treated already and have been for centuries. They are taught the limits in life, they are taught what is right and wrong and the parents monitor them until they can act responsibly. This is not much different, with the exception that the sending & receiving of images would be flagged up to parents automatically, because it is no longer feesible to monitor what the kids are doing on their devices, unlike earlier, where they weren’t online or the devices were in a common room, where the parents could keep an eye on them easily or put parental blocks on the devices and the network connection to stop them visiting known bad sites.

Social media and mobile devices that are outside of the home network have moved the goalposts and made it nigh on impossible to monitor what the kids are doing.

I don’t mean that parents used to monitor every site they visited or every floppy disk or cassette the kids brought home with them, but we were still taught what is right and wrong. I think the Apple system of alerting the child that the image they are about to receive might be pornographic & do they want to still receive it, likewise, refusing to send naked images from their devices, is a good one & the parents are warned if the child goes ahead and breaks the warning.

There is trust there that they will do the right thing & their parents will never find out and there is control, in that if they decide to still break the “rules”, the parents will be informed.

I am going through this control at the moment, not with graphic content, but with teaching rules and consequences, we have my daughter, her husband and a 2 year old and a 5 month old living with us for the last 6 months. It is a daily battle to show the 2 year old what is right and what the consequences of doing the wrong thing are. She even knows a lot of what is wrong, but is continually pushing those boundaries, “just how far can I go, before the big ones start taking away my toys/food/drink”?

I agree, this doesn’t cover the other side of the problem, adults sharing with adults. But that boat has long since sailed. The encryption is out there and you only catch the idiots today, who distribute it on public networks, like Facebook. Most probably are already on the dark net - there have been several cases in recent years, where hundreds of people have been arrested, because the authorities found the dark net server, infiltrated it and followed the users back to their devices from there. This shows that, for the hard core perpetrators, they already have the means to find them and deal with them and the small fry are caught through their own stupidity. That leaves some in the middle, who are dumb enough to use public tools, like Telegram, WhatsApp, Signal etc. but clever enough to choose one that offers E2EE.

They are proposing that the end to end encryption be broken and that a third party with the back door key can listen in centrally to the conversation. The problem is, they don’t understand how encryption works, or that history over the century, heck even the last 20 centuries or more has proven that having a backdoor to encryption just doesn’t work.

Once you have a way to break into the communication, “everybody” will do it, jealous partners, criminals etc. as well as the “key holders”, E.g. the police.

You and I must have had very different upbringings. I was running a $100/week paper route when I was 8 years old (until I was 16.) My parents operated on the assumption I was doing the right thing or else they would hear otherwise. Since I had great school grades and a 6 day a week paper route, and my computers, I was probably too busy to find any trouble. I do realize times have changed somewhat, because back then, when I was at least 6 years old, I was walking to school, biking in the neighbourhood, and going to the local swimming pool in the neighbourhood park. I don’t think, back then, we lost sleep over “stranger danger.”

I didn’t necessarily mean at older ages, as you say, times have changed. But as a small child, you are monitored by your parents (and later, teachers) and taught how things (society) works and what is acceptable or not and, once you can prove you understand the differences, controls are gradually eased back, until they are no longer needed.

It is just that this control is, by the nature of social media, puberty and other factors, happening later than most other control and monitoring. This is also why I think the Apple system is not too bad, as it chides the teenager, if they try and send a nude picture, but doesn’t tell the parents unless they ignore the warning. If the warning isn’t enough to make the child think twice about sending a possible nude of themselves, maybe the parents do need to have a word with them.

I used to ride over to my uncle’s farm (15 miles away), when I was a kid. I walked to school as well and I had a saturday job, which paid 4 quid a week, from 13 through 16, repairing and preparing old computer hardware for a company that bought liquidated assets from failed companies.