On Mind Control, Part 2: Word-to-Vector

Written by Sasha Latypova on August 22, 2023. Posted in Current News

Have you ever wondered why liberal arts were called that. Liberal?

The words stem from liberty. The term “liberal arts” for an educational curriculum dates back to classical antiquity in the West.

The seven subjects came to be divided into the trivium of rhetoric, grammar, and logic, and the quadrivium of astronomy, mathematics, geometry, and music.

Before they became known by their Latin variations the liberal arts were the continuation of Ancient Greek methods of enquiry that began with a “desire for a universal understanding”.

In 4th-century-BC Athens, the government of the polis, or city-state, respected the ability of rhetoric or public speaking above almost everything else.

Originally these subjects or skills were held by classical antiquity to be essential for a free person (liberalis, “worthy of a free person”) to acquire in order to take an active part in civic life, something that included among other things participating in public debate, defending oneself in court, serving on juries, and participating in military service.

The language-based skills of public speaking, writing and logic were established in the Western culture as an attribute of a free person, in contrast to a slave, serf or a servant who would be deficient in those skills.

Critical thinking is a skill which is driven by the cycle of grammar, logic, and rhetoric. The purpose of grammar is to bring a consistent order to a body of knowledge. The purpose of logic is to extract understanding from the body of knowledge.

Rhetoric is the cogent explanation of that body of knowledge. In ancient times, students were taught the Trivium at home, by their parents, as a pre-requisite for admission into universities.

Grammar-Logic-Rhetoric, Grammar-Logic-Rhetoric, etc. = a comprehensive decision making process. When this human learning process is malformed, stunted or deficient in some way, we cannot arrive at the truth by critical thinking process, and instead accept whatever feels emotionally pleasant as truth.

We emote instead of thinking. Intellectually, emoting is a lower level space which leads to the mind enslavement, passing the operational control from self to others.

Given what we know about the state of affairs in our world, it should not surprise any of my readers that the Trivium is not part of the required curricula in the public schools or universities today.

The public education system is not interested in educating free citizens, they are a factory for churning out standardized, emotive and obedient cogs instead.

It is no wonder that 24 centuries later the AI developers concern themselves with language based models. Also no wonder that those who seek to control and dominate us, start by taking over, controlling and dominating our language.

It starts with political correctness first, proceeding to censorship of the language, to rabid woke ideology, then outright criminalization of unpopular or dissenting speech. All with the help of AI and surveillance.

Let’s look into a very interesting category of AI training models called Word Vectorization (or Word Embeddings).

Word-to-Vector AI Models

Word-to-vector is a technique for natural language processing (NLP) that learns word associations from a large body of text such as text available on the Internet. Text scraping tools are used extensively by both the intelligence apparatus and AI developers for all kinds of applications, most of them for bad purposes.

Let’s just focus on the theory for now. The idea is to represent each word mathematically by a vector of real numbers that captures the location of the word in relation to all other words. During training, the model reads through huge amounts of text and sees how often words occur next to each other.

Each time words occur next to each other, their locations in vector space is nudged closer.  When they are far apart, their locations are nudged further away. This process is repeated millions of times until all the text has been processed.

The result of this massively repetitive process is a map of words in a vector space with their locations. Words that are similar to each other are located close to each other, and words that are very different from each other are located far apart from each other.

Mathematically, words that are similar in meaning or usage would have similar vectors, and words that are different would have different vectors. This way, you can measure the similarity between words by using a simple mathematical function.

Below is a simplified example of countries and cities mapped onto a two-axis vector space (completed through principle components analysis) that was created by analyzing large volumes of text from the Internet (Source: A Beginner’s Guide to Word2Vec and Neural Word Embeddings | Pathmind).

As can be seen, the countries are clustered together and the cities are clustered together. Two countries that are similar to each other in the real word (Germany and France) are also close to each other in vector space.

Germany and France are both large European countries with large economies and populations which are located close to each other.  And naturally, Germany and France are closer to each other than they are to Japan which has a very different culture and is far away from Europe.

Same behaviors can be seen for the cities. Madrid and Lisbon are more similar to each other as sharing similar languages (Spanish and Portuguese) in Europe than they are to Tokyo which is on the other side of the world with a different culture and language.

One of the most striking discoveries of Word-to-Vector modeling was the realization that the vectors could represent much more complex relationships. If you look at the graph above, all the arrows from the countries to the cities move from the left to the right at approximately the same angle and for the same distance.

In other words, these arrows all represent the concept that the item on the left is a ‘country’ and the item on the right is a ‘capital city’. Specifically, the following holds true for each pair:

Rome – Italy (i.e, arrow from Italy to Rome) = Beijing – China (i.e., arrow from China to Beijing)

This allows for very intriguing math to be completed between the vectors.  If you wanted to find out what the capital of China is, you could simply calculate the following.

Rome (capital of Italy) – Italy + China = Beijing

You calculate a number of other curious equations that yield interesting and intuitive results.

Home Depot – Man + Woman = J.C. Penney

Doctor – Man + Woman = Nurse

Iraq – Violence = Jordan

President – Power = Prime Minister

Human is a Living Language

Note: the whole Universe is a living language, and in my opinion it is God’s spoken word through time and space, but we will examine this in another episode.

All of us have our own word-vector spaces in our heads. As we hear words and read text throughout our lives, our own mental representations of the vector space develops, adapts and changes. People belonging to same cultural, generational, professional or social groups have similar word-vector spaces.

There are specialized spaces, such as slang, or ‘cant’ (inside language) designed for shorthand and/or to exclude outsiders. After we acquire language fluency, what happens most of the time – we are simply running on a word-vector autopilot.

Words most commonly used together, reside in the vector space close to each other and get reproduced together in common sequences (phrases) and we don’t spend any time searching for them.

Most common daily speech is auto-generated without much thinking, it only requires some initial prompts – key words from which the autopilot will string the expected phrase of closely-vectored words. Only some of the time the “operator” level thinking intervenes and overrides the auto-pilot.

Human as Trinity: 2 minds + 1 body

This gets into the theory of dual mind, which goes back to antiquity as well. The idea that a human is 2 minds + 1 material body (a trinity). The two minds are called different things, sometimes it is soul and spirit, sometimes intuitive and rational.

There is an interesting book “Thinking Fast and Slow” that describes some of this (but not all). I am going to call it “autopilot” and “operator”.

Autopilot or intuitive mind controls autonomic functions of the body, and is amenable to control by suggestion when the operator is not on guard or willingly surrendered. This explains the entrainment success for some people discussed in Part 1.

The “autopilot” mind contains all information acquired through the lifetime, but not statically – it is constantly shaped, trained and redeveloped through daily experiences of the “operator” or rational mind.

This is neither good nor bad, it just is. It is a very energy-efficient way to use the brain (which consumes by far the most amount of energy in the body). For example, you can prepare for job interviews by acting out various interview scenarios.

This is a way to train the word-vector autopilot space, which reduces anxiety and builds confidence for the actual job interview. On the other hand, next time you are in a “small-talk” situation, just observe how automatic it is!

Thus, I dislike small talk as a waste of time. With a spouse or a close friend, you can often “read their mind”, and complete their sentences, not because of telepathy, but because there is a substantial overlap between yours and their word-vector spaces (i.e., both autopilots trained through substantially same experiences).

However, lack of awareness of the natural language processing (by human brains or by AI) can lead us into trouble. I discussed in Part 1 that humans, as all live things, are varied.

Your vector-space is different from mine, it is also different from your space a year ago, and it also varies by time of day, by your location, mood, activity and other circumstances. The computer AI models are static and always the same.

That’s why they age and decay fast and need to be constantly updated.

The danger is that most people do not know that most of their daily conversations are generated by their own word-vector autopilot. Thus, the autopilot is subject to capture and re-training by non-friendly forces, and surprisingly little technology is required for this.

Social media platforms sell your data to “scrapers” which use it to train and updated AI models. The largest scrapers are the government-corporate cartels that use this for nefarious purposes – censorship, fake social media accounts, better hacking tools, fake news generators, “prebunking narratives” and many other informational warfare methods, which are largely all linguistic.

Language as a Weapon of Enslavement

It can be argued, that Word-to-Vector is able to represent our verbal understanding of the world and could be used to manipulate people by changing their mental models (or their own internal word-vector spaces).

Indeed, this is the nature of the 5th gen wars. The war is waged in the mind and in the word-vector space first. For example, in the past the words ‘transexual’ and ‘mentally ill’ were located close to each other in the mental models of most humans and in the Word-to-Vector space.

However, if you repeat the following three words close to each other over and over again: ‘transexual’, ‘rights’, and ‘gender affirming’ the mental model is changed. This same process can be applied to vaccines by repeating over and over the three words ‘vaccine’, ‘safe’, and ‘effective’.

In other words, there is no need for advanced technologies, brain chips or injected MAC addresses to control the minds of others! Simple repetition of words and phrases will brainwash most humans if they are not paying active attention or do not have skills to recognize and counteract this attack on their minds.

Once you realize this, you can see this technique being used everywhere.

This is the principle behind Operation Mockingbird – repetition of simple phrases to the audience whose critical thinking skills are dulled or destroyed by bad education, encouraged disorganization, consumerism, addiction to TV, wokeness, bad nutrition, “public health”, the CDC’s childhood poisoning schedule and media propaganda.

No need to use any brain implants or nanotechnology. No need for high-tech transhumans. They are already here.

The merging of human with machine interface is accomplished by “casting spells” on humans that, conveniently, have never been trained in critical thinking, and have never been educated beyond the level of a compliant obedient serf.

This is taken from a long document. Read the rest here substack.com

Please Donate Below To Support Our Ongoing Work To Defend The Scientific Method

PRINCIPIA SCIENTIFIC INTERNATIONAL, legally registered in the UK as a company incorporated for charitable purposes. Head Office: 27 Old Gloucester Street, London WC1N 3AX. 

Trackback from your site.