Bots, book and Bible ciphers, Want Ad secret messages—and Twitter posts (X). How do these things interconnect?

A breakfast-time talk about an odd Twitter post that I stumbled onto led to me teaching Emily about book ciphers, want ads, and one-time-pads. OSINT (Open Source Intelligence) meets OSCRYPT.

Before I begin today’s post: please see a personal note from me to all of you at the end of this piece—and thank you for being a loyal reader.

---

Now for today’s post.

If you're a regular reader of CognitiveCarbon, you know the pattern: I begin an essay by taking you on a journey, starting out somewhere mundane—but ending up somewhere entirely different. Along the way, the story may teach you new things, viewed from a perspective that perhaps you hadn’t had before.

If you enjoy this style of mine, read on! (and consider becoming a paying subscriber. It is genuinely needed and helpful.)

Today’s topic began over pancakes this weekend; sometimes the best inspiration for a piece originates from a humble and mundane spark.

To set the frame: there are covert actions going on in our world that we can only dimly perceive. Surely this isn’t a surprise to you. As I mentioned in a previous post about AI (Assistive Intelligence), however, there may soon come a day when beneficent AI helps us pierce through the fog, lies and deception that wafts like a sickly cloud over us all.

This kind of useful AI will broaden our knowledge and enhance our understanding of the truth—potentially helping to reveal the true nature of the world as our AI assistant helps peel back layers of purposeful deception and distraction.

At breakfast on Saturday—while we were munching on my specialty pancakes with bits of crunchy bacon hidden inside— I was talking with Emily about some strange posts that someone had surfaced the day before on Twitter.

I immediately recognized the pattern as a possible instance of OSCRYPT: i.e., secret message passing using “open source” channels.

After I explained to her what might be going on, and gave her the background knowledge, several things connected for her that she had long wondered about—but lacked the words or concepts to fully grasp or express.

A lightbulb went on for her; but along with her new awareness came a new source of worry.

Because she got so much out of this Saturday breakfast chat, I thought I would share this topic with a wider audience—in case it helps you, too.

OSCRYPT: what exactly is it, and what might it indicate is happening?

You can think of OSCRYPT as a form of “open-source cryptography”, which is a phrase that I personally use here to mean “secret message passing out in the open.” There are other uses of the phrase, but this is the way I’m defining it for today’s post.

OSCRYPT means that anyone can see a message (with hidden context) in plain sight in some public forum or channel, but only someone with special insight can extract its meaning.

Some of you may be familiar with the term “Steganography”, which is an analogous idea.

From Grok on X:
Steganography is the practice of hiding a secret message, image, or file within another message, image, or file. It is a form of covert communication that can involve the use of any medium to hide messages. The term "steganography" comes from the Greek words "steganos" (meaning "covered" or "hidden") and "graphia" (meaning "writing").

Steganography is different from cryptography, which is the practice of protecting the contents of a message by encrypting it. In steganography, the goal is to hide the very existence of a message, making it undetectable to anyone who is not aware of its presence.

There are many techniques for hiding messages using steganography, including hiding data in images, audio, video, and even text files. For example, a message can be hidden in an image by manipulating the least significant bits of the image's pixels to represent the message.

Steganography has been used for centuries to conceal secret messages, and it has become increasingly important in the digital age as a way to protect sensitive information.

In this case, what surfaced was a series of tweets (X-posts) using identical wording that were posted by several seemingly unrelated accounts. At the time, there were four copies of this:

This pattern of “identical tweets” is familiar to many of us who have been learning OSINT techniques because of our interest in Q; we’ve seen it used over the years as a means to engage in divisiveness attacks (where a boilerplate reply is designed to irritate or enrage a group of people so as to fragment them, demoralize them, or dilute their message) or else as a not-very-skillful attempt at echo-chamber narrative steering.

It often relies on automated Twitter/X accounts (so-called “Bots”) which inject the identical message many times over.

The frequency of this exploit has been substantially reduced in the last year, thanks to the moves made by Elon Musk to make the creation of bot accounts more expensive. But that change still leaves another technique unaddressed: cracked passwords on existing accounts.

To see this sort of thing in action, for example, you might see a bunch of X accounts with a leftward ideological tilt all trying to spam various posts and threads with the identical copypasta reply message in order to memetically inject a particular narrative.

Sometimes it’s just teamwork by a group of leftist radicals; other times it’s automated.

What is unique about this particular ‘bot message’ was that it didn’t have an obvious purpose of inflaming, attacking or dividing; and it didn’t have an obvious purpose for narrative pushing.

It was just plain…weird.

There were some odd juxtapositions of words, and one which was oddly capitalized. At the time, there were four different unrelated accounts that had posted this message. When I went to search for them later, all but one had been deleted. I thought I might know why.

So what might be going on here?

Let’s take a step back many decades in time to learn about an old technique of spy craft, and then explain why one might want to use an “old school” technique like this today—even though there appear to be many messaging options that (seemingly) provide “secure communications” with end-to-end cryptography (e.g., iMessage, Signal, Telegram, WhatsApp, etc.)

Even if these platforms actually provide secure communications (meaning that the message legitimately can’t be intercepted and decrypted) there are still problems with using them.

Bill Binney, formerly of the NSA now turned whistleblower and the inventor of ThinThread, talked about in this documentary about him:

The problem is that, as Binney explains, you don’t really need to decrypt the actual message to still get a gist of what is going on.

You just need to keep track of who talks to whom, when, how often, and then track who the first recipient talks to next, and so on: and then watch carefully for evidence of what happens a short time thereafter. (You monitor ‘Hops’, in the parlance of the detestable FISA.)

If you keep track of just these things, you can often infer what the message might have been (in fact, this technique has also been used in some form to go back later and ‘crack’ the encryption used for the original message; once you know the outcome, you can try to crack the codewords that led to it.)

If you are aware of this fact—that even secure message platforms can expose your hidden intents—and you were a covert operator, you would avoid these channels altogether (not to mention that they might have backdoors for the NSA to decrypt your message, anyway.)

In that case, you’d choose a simpler and more foolproof method for “hiding your message”, similar to book ciphers or one-time pads. Before we get to explaining those things, let’s take a Chat-walk-A (Chautauqua) back in time.

When I was growing up in the 70’s and 80’s, I enjoyed reading “spy novel” fiction and watching similarly themed movies and TV shows. I learned how spies who were sent to the US from a foreign adversary might go deep undercover for years or decades, waiting for a signal from their controllers back home to begin some operation.

Some of you already see where this is going, don’t you.

Because the use of radio signals to communicate between the ‘handler’ and the ‘operator’ might be detected and give away information (at the very least, the origin of the signal sent by the controller, and the timing of the signal) the spies would use a different technique: they would place ads in the want ad sections of large city newspapers offering to buy or sell something.

For those of you too young to know what “want ads” are: think Craigslist, but in newspaper form. In the old days, people would post short one- or two-line ads for things they wanted to buy or sell, paying their local newspaper a small fee to include their ad for a few days or weeks.

Some clever person realized that because newspapers often had a large geographic reach—and provided an essentially anonymous way to “post a message”—they could use the want ads as a messaging channel, by including special keywords in the post.

These keywords would mean nothing to the public at large; but for the sleeper cell spy waiting for instructions, the keyword might inform him of the time to begin or give him the identity of his target.

The words used would only have meaning to the sender and recipient; to the rest of the public, they thought, for example, that they were just reading a want ad like “wanted to buy: 1930s era teddy bears with polka dot neckerchiefs”.

But these words might have told the receiving spy to begin his covert operation.

Various forms of “secret messaging” have been used for thousands of years—since the time of Sun Tzu—to covertly communicate instructions to militaries or operatives.

One technique that arose involved the use of well-known books, including the Bible. Only the secret message sender and his recipient knew the “code”.

The sender would look up words in a certain version of the Bible, for example, within a certain range of pages, and send just the page numbers, line numbers, and word numbers in a sequence.

The recipient would see, for example, “101, 12, 2” and would go to page 101 of his King James Bible, find line 12, and then find the 2nd word and write it down.

After doing this laboriously for all the other sets of numbers in the secret message, he could reconstruct the original message (“meet me by that one wall at the special place on this date so we can do the thing”.)

If you are an astute observer, you’ll notice that something similar to a “Book Cipher” was used by whomever was behind Q; they used a corpus of related and interconnected “messages”—that anyone in the public could see—to send “secret” clues (using references to times and drop numbers) to those in the know. All with plausible deniability.

Knowing this fact, among other features, is what made Q highly credible to me. Q has a certain…rationality to it.

Over the millennia, the arms race between the “message encryptors” and the “code breakers” would lead to more and more sophisticated techniques to hide a message in plain sight.

One well-known technique—that even today is still essentially unbreakable—was used during WWII, the Cold War, and at other times; it involved so-called “one-time pads”.

A one-time pad (OTP) contains an essentially random arrangement of “plaintext” letters and their scrambled equivalents. To encode a message, you ‘translate’ each original letter of the message by looking up its replacement using the pad to transform the whole message into its scrambled variant; then you send the scrambled message (which looks like gobbledygook to an interceptor) over your open channel.

Modern digital cryptography does something similar, using a mathematical algorithm to encode and decode the human-readable message. But the internet as we know it has only been around for a relatively short time in the history of secret communications.

During WWII, they’d use morse code and radios to send covert messages using OTPs. The recipient, having the same (but reversed) version of that one-time pad, could then reverse the translation to retrieve the original message. The people eavesdropping on the radio or morse code would only see nonsense.

With this technique, a set of shared ‘code books’ would be held by the originator of the message and its recipient, maybe in a safe, and after a particular message was sent and received, the page used would be "torn out and discarded” and never used again.

This technique is essentially uncrackable, provided that certain conditions are met. The only way to break this kind of code is to covertly steal the sender or recipients’ codebooks.

To protect against this, however, agents would only keep a certain number of days’ worth of pads, so that even if these were stolen and used to intercept messages, once they ran out then “fresh pads” could be produced to re-secure the message channel.

These OTPs were used up to the modern day, including in nuclear missile command and control.

If the Germans had done this instead of using the Enigma machine during WWII, then the codebreaking machine—which Alan Turing, the father of the modern computer helped to design at Bletchley Park in the UK in order to crack Hitler’s code—might never have been able to intercept Hitler’s messages to his generals, and the outcome of the war might have been entirely different.

I wrote about Alan Turing in my substack post about consciousness a few years ago. If you are fascinated by this phase of history, watch the movie The Imitation Game and read the book Cryptonomicon by Neal Stephenson.

So now let’s return to the mysterious Twitter messages.

You can now understand that these, too, could be a form of “book ciphers”, in which various words (the words themselves, their arrangements relative to each other, and things like punctuation or capitalization) might encode certain information intended for sleeper cells.

If you were to use such a technique on Twitter/X, you would want to hide your actual identity—you wouldn’t send these from your own account.

Thus the use of either bot accounts created by anonymous means, or the use of compromised accounts whose passwords you were able to crack (or for which you perhaps purchased the passwords on the Dark Web), which you “kept in your back pocket” for the right time to use.

In this way, for example, an innocent and unknowing person might have their X account used to (very rarely) send a reply to a tweet with a “coded message”. The recipient, having seen the message, could then post a follow up message in a tweet indicating “received”.

Seeing this “received” message, the original sender could then go delete the copies of his reply tweets and erase all traces of his actions. If “caught”, an innocent person might receive the blame, but because the message seems innocuous, there is also plausible deniability at work.

In fact, the original, innocent account owner might not have even been aware that a message was temporarily sent from his/her account and then later deleted.

Why use multiple copies of the same message? For two reasons: to make sure the intended recipient(s) would see it, to obscure the original sender behind network complexity (find Waldo in a sea of people wearing red and white shirts), and to ensure that if a compromised account owner somehow finds the injected message and deletes it, that there are still other copies floating about.

The victim account holder, seeing a nonsense message appear in their replies, might not even been concerned enough to report it (“huh. what a weird glitch. AI these days!”)

If you think back, now, you might remember hearing news stories with the notion that “the Intelligence Community has detected elevated levels of chatter” during some period of geopolitical unease, warning us to be on alert for some significant attack or bad event.

What this likely means is that these so-called “intelligence” agencies (a term that I deconstruct in this humorous but biting post about UFOs and the Royal family) are at least aware that the quantity and frequency of (for example) “anonymous OTP or book cipher messages” has increased noticeably and have been detected, even though the agencies don’t—and can’t—know what the message might be.

All they can say is “the noise level has increased as of late”, so be on alert.

So, to finally close the loop: appearance of odd and seemingly random messages like these on Twitter/X or Facebook might mean that sleeper cells are being readied for action. What action, and which sleeper cells? We can only guess. Forensic study of social network traffic might yield some clues.

Can Twitter itself at least detect, algorithmically, the existence of these kinds of patterns because of the message duplications and subsequent deletions? It certainly could; and for this, and many other reasons, the FBI and the CIA might have an usually large interest in “Twitter” and who owns and operates it.

Because it’s a valuable tool to them, both for deceiving the masses though memetic injection and narrative steering, but also for detecting “chatter”.

Now you know the rest of the story. Keep an eye out for things like this on Twitter/X or other platforms—and share examples that you find with me in the comments here.

---

Now for a personal note.

Next month marks the one-year anniversary of my move to Michigan from California, which had been my home for the prior 34 years. I’ve spent much of the last year building a new life and home around Emily and Evelyn, who are my purpose for making this life transition.

It’s been a year of nearly incomprehensible change for me, even as the world itself transforms at an even more dizzying pace—what with the advent of AI and the growing Great Awakening among a larger slice of the population following the great pandemic “control experiment”.

Living here in Michigan, I’ve begun what feels like the next of the many lifetimes I seem to have already lived through in my time on the planet. God has a reason for putting all of these experiences before me, I’m sure.

Over the last year, I’ve drafted many posts here on Substack that I didn’t have time to complete. This week marks a return to my prior pace of posting, with new thoughts and ideas on many topics to share with you. I’ve spent a few months taking a deep dive into AI as I’ve tackled a new part time job, which has been immensely valuable.

I’ll release four new posts over the coming week, starting with this one—with more to follow next week.

Thank you all for being loyal followers—your presence here in the CognitiveCarbon zone makes you family of a certain welcoming sort, and your thoughtful comments are the jet fuel that keeps me going.

—CognitiveCarbon

---

I hope you enjoyed this latest post! More to come soon on AI, the Pandemic, and other topics.

CognitiveCarbon’s Content is a reader-supported publication. To support my writing and research work, please consider becoming a paid subscriber: at just $5 per month, it helps me support a family.

You can also buy me a coffee here. Thank you for reading!