# A cryptographer’s thoughts

Aug 9, 2023

An explanation of the numbers and the need for severance from a cryptographer.

I’ve spent the last two decades working (and playing) in a variety of fields, including software engineering, cryptography, and security. This is what I believed the refiners were doing right away after watching this incredible mind-bending show for the first time. I might be absolutely wrong and I might be viewing the performance through professional lenses, but after reading The Lexington Letter and other information supplied here by Dan Erickson, the notion seems to be supported.

You use asymmetric encryption (and with a few other things) to keep your shit locked up every time you visit a website that begins with “https” rather than just “http”. There are several varieties; two well-known ones are RSA and Elliptic Curve. A computer can multiply a lot of (relatively) little integers to create a huge one reasonably easily, but it is virtually impossible for a computer to take that enormous number and determine what the smaller ones were once more. This is how RSA works. Elliptic Curve functions because it is incredibly simple for a computer to predict how a ball would bounce off a number of hills, but nearly impossible to determine where the ball ended up and figure out the shape and position of the hills it bounced off of.

In fact, that is the fundamental principle behind all asymmetric encryption: You need an issue that is incredibly hard to solve yet relatively simple to validate the solution to. This is the “NP” or “NP Complete” problem that math nerds always refer to. Right now, if I handed you a finished Sudoku puzzle, you could probably give it back in a moment and tell me if the answer is correct or incorrect. However, doing it alone would take a lot more time. It would take you proportionately longer to confirm the solution is correct but frigging ridiculously EXPONENTIALLY longer to come up with a solution if I added a lot more numbers and gave you a Sudoku that was, say, 64×64 instead of 9×9.

Look, there’s a good likelihood that “NP Complete” problems won’t ever have a rational solution. But consider how these issues may be solved if we had a different kind of computer. They couldn’t logically solve the extremely large Sudoku puzzle number by number, but what if this computer could concurrently consider every potential combination of solutions, correct or incorrect, and then check each one to determine if it’s the right answer? The incorrect answers would disappear, leaving only the right one. Although greatly oversimplified, this does exist. A quantum computer is already here and exists; it is still in its very early phases and requires further research before it can be used, but we already know that one day in the near future.

It will outperform RSA encryption. According to some experts, a quantum computer could decrypt any asymmetrical encryption now in use.

There isn’t much coverage on this. The simplest explanation of why we’re in trouble can only be gained by reading a number of articles, and that doesn’t fit into a 20-second news clip. Although this issue has been well-known for DECADES, it has just recently started to materialize in recent years. Effectively, quantum computers are able to analyze all possible solutions to a problem at once and determine which is correct by logically comparing them all. This technology will render useless any system that can currently perform the calculation. If such takes place, your private

Your entire digital existence, including your bank account and images, is available for download. This won’t happen in the distant future; it will actually occur in our lifetimes.

Scary, huh? It appears that we require encryption that is not reliant on a computation issue. But what could you possibly use when everything, even music and fingerprints, can be quantified? As it turns out, the smartest and most sophisticated AIs we now possess are wholly incapable of experiencing human emotion. What if our new encryption relies not on factoring numbers or bouncing balls, but rather on people experiencing emotions in response to the data as the fundamental hard-to-solve, easy-to-verify problem? Although a quantum computer might be able to imagine any possible arrangement of numbers,

Never be able to physically verify any of the infinitely potential answers.

You object, This is absurdly technical and not something this show would ever subject its viewers to! Although intriguing, this show is in no way based on it.

Well, definitely, sure, of course, I respond. However, unless they want to plunge in, viewers are not required to know this history. “Our [government/competitors/political opponents/enemy] was able to procure a computer capable of breaking all codes that a machine can generate,” is all that the average viewer needs to know. Because the human brain can feel emotions and computers cannot, we have integrated it into our machines.

But what gave rise to this hypothesis? We might, after all, assume that these figures are for anything. Why do I have such a strong feeling that our four refiners are today’s equivalent of Alan Turing’s device that deciphered the Enigma code during World War II? What is the REAL cause for this employment to necessitate severance? is the first question you should ask yourself.

A brain implant may be necessary to have real human emotions respond to numerical matrices, but why would this affect memory? Work-life balance is undoubtedly a subpar excuse. And to be completely honest, I don’t buy that it’s used to safeguard business secrets; you can achieve the same goal via rewards and penalties without investing a fortune in memory suppression technology. Perhaps it’s due to the nasty nature of the narrative

Although I don’t buy it either, rewards and punishments wouldn’t keep employees quiet. If/when individuals become aware that their effort is being exploited for a malicious purpose, they could just quit working, repercussions notwithstanding. No, I believe the severance procedure would only make sense if both the innie and the outie were exposed to independent pieces of knowledge that, when combined, would betray the work’s secret.

The Lexington Letter enters the picture at this point. If you haven’t yet, you should download that for nothing from Apple Books. It’s an easy read and canon, in Dan Erickson’s opinion. The “innie” At 2:30 pm, Peggy reports to her outie Peg that she completely solved a file. The outie is aware that at 2:32 pm, a vehicle belonging to Lumon’s main rival detonated. The only reason anyone was able to connect those two occurrences is that Peggy and Peg were able to speak with one another after Peggy cracked the encryption key and Peg saw the actual results. This is the reason why these people need to be cut off.

This is the reason why these people need to be cut off. everything wouldn’t take long for code breakers to piece everything together if they observed significant global events happening minutes after each file was finished. For the same reason that Turing’s codebreakers at Bletchley Park had to consent to German attacks even though they were aware of them beforehand. Nobody could know they made the connection, therefore they had to keep it a secret.

But this episode has a lot going on. Let’s examine the details:

You’ll need to conduct a lot of research on emotional triggers if you’re creating an emotion-based encryption (or an assault against one). You would require, say, a department of art that circulates artwork around various audiences and gathers comments on it. Studying how parents react to a baby’s cries or raising infant animals would be necessary. You may keep a big database of songs from different genres and track how people react when given the chance to dance and listen to them for brief periods of time. Heck, be lazy and write emotions like “Defiant” right on them; individuals with serious mental disabilities won’t understand. Perhaps you disguise that as a “reward” so the participants won’t see it

Heck, be lazy and write emotions like “Defiant” right on them; individuals with serious mental disabilities won’t understand. To keep the subjects from realizing they are being used in an experiment or training program, you might even pass it off as a “reward”. When confronted with a variety of emotional stimuli, you may occasionally send employees to a room to see how tough it is for them to control their emotions. Call it the “Wellness room” to make it sound beneficial as well.

Files go away. No matter how much time and effort is put into them, if they are not solved by a particular time, the refiners start again with a fresh file. As in World War II, when the Germans would alter the code for the Enigma machines every day and the codebreakers at Bletchley Park had to crack it before the change, failing which they would have to restart with the new code the following day.

Woe, frivolity, dread, and malice are the four temperaments. It has already come to light that the refiners also label their numbers in this manner. Here, we’re dealing with both computer algorithms and human emotion. There must be no room for error or overlap in this. These four words represent extremes, making it impossible to mix them up. Although “Anger” and “Sadness” could be interpreted in different ways, these four emotions are all located on opposite sides of the room. There is no chance that one might be confused with any of the other three.

Breaking extremely sensitive codes makes you a target for cyberwarfare. Utilizing exclusively non-networked workstations, or at the very least machines that do not support all of the network features offered by contemporary operating systems, is the greatest approach to assure that you cannot be hacked. equivalent to, say, a really outdated computer that can only support a garish green monochrome display.

But this can’t be scaled, is it not? We have four complete persons here, after all.

Here’s where I start to speculative. Refining may serve as a training ground for a machine learning algorithm, as has previously been discussed. That might be true, but in my opinion, investing in a brain implant would be pointless if the problem could be resolved by machine learning. It would be absurd to market that implant to the broader public. That doesn’t seem to be the entire narrative.

I believe that deciphering codes requires a human brain that is capable of feeling emotion and always will. Perhaps, though, it won’t always involve individuals sat at computers. Maybe effective software will simply need to communicate with a brain through an implant in order to function. Scaling up then merely entails placing an implant in as many brains as possible. It now makes sense to sell this to the general population as a tool for improving life. Now that we’ve made everyone want it, we have a sizable network for cracking codes. Since Lumon is now the only entity capable of decrypting post-quantum period encryptions, neither corporate nor international government secrets are hidden from them. absolute dominance.