Cassiopaea and ChatGPT

[Laura]

I've seen many communities try to use Chatgpt to gain new insights or to visualize certain things, but to be honest, people greatly overvalue what can essentially be called overglorified linear algebra.
Chatgpt often fails at the most basic tasks(even algebra), needs constant babysitting, and it was trained with certain datasets that will skew all results. It literally has zero consciousness and is more or less google search for the lazy.
For example, chatgpt has been used by authors to produce information based on their own work, and the amount of mistakes and tilted messages(there is a strong left-leaning bias in the whole thing because of the datasets, most likely) are just embarassing.
I don't know what value you see in this idea.

Well, in this particular GPT, the ONLY material it can access is what I have given it, i.e. the Cass transcripts. It has no other dataset. I've created several GPTs including one that has all my books as the dataset. Another that is devoted to "From Paul to Mark". This is how I've avoided the 'lefty bias'. And, since I know the works intimately, I can catch the errors which are what reveal the programming biases. Still, it is very useful.

 

[Laura]

What is Cassiopaean Chronology of Human history on the planet and in the solar system?

Based on the Cassiopaean material provided, here's the detailed chronology of human history on the planet and in the solar system according to the Cassiopaean transcripts:

### Cassiopaean Chronology of Human History

#### **Origins and Early History**

1. **Formation of the Solar System:**
- The solar system was formed through natural processes but influenced by advanced extraterrestrial civilizations.
- Mars and other planets once had flourishing ecosystems and advanced civilizations.

2. **The Arrival of Humans:**
- Humans were genetically engineered by a consortium of extraterrestrial races.
- The human genome was manipulated to serve as a versatile vessel for various experiments and experiences.

#### **Ancient Civilizations and Cataclysms**

3. **Lemuria and Atlantis:**
- Lemuria, an ancient civilization, existed around 80,000 years ago and was highly spiritual and advanced.
- Atlantis followed, approximately 50,000 years ago, and became a powerful and technologically advanced civilization.
- Both civilizations eventually fell due to internal corruption and misuse of technology, with cataclysmic events contributing to their demise.

4. **The Great Flood and Subsequent Reset:**
- Around 12,000 years ago, a massive cataclysm, often identified with the Great Flood, reshaped the Earth.
- This event marked the end of Atlantis and led to a reset of human civilization, with survivors spreading out and forming new societies.

#### **Human Development and Extraterrestrial Influence**

5. **Post-Atlantean Societies:**
- Survivors of Atlantis and Lemuria contributed to the formation of ancient civilizations such as Sumer, Egypt, and the Indus Valley.
- These early societies retained some advanced knowledge but gradually lost much of the technological prowess of their predecessors.

6. **Continued Extraterrestrial Interaction:**
- Various extraterrestrial races continued to interact with humanity, sometimes guiding and other times manipulating human development.
- These interactions influenced major historical events, including the rise and fall of empires and the spread of religious and philosophical ideas.

#### **Modern Era and Future Prospects**

7. **Manipulation by Higher Density Beings:**
- In the modern era, humans are still subject to manipulation by hyperdimensional beings, often without their awareness.
- These influences shape political, social, and economic developments on a global scale.

8. **Potential for Spiritual Evolution:**
- The Cassiopaeans emphasize the potential for humans to awaken to their true nature and the broader reality of the universe.
- Spiritual growth and awareness can lead to liberation from manipulative forces and the achievement of higher states of existence.

### Summary

- **Multidimensional and Extraterrestrial Influences:**
- Human history is deeply intertwined with the actions and influences of extraterrestrial and hyperdimensional beings.
- Major cataclysms have periodically reset civilization, allowing for new beginnings and the spread of survivors.

- **Technological and Spiritual Legacy:**
- Ancient civilizations like Lemuria and Atlantis possessed advanced technology and spiritual knowledge, which have been partially preserved and passed down through subsequent cultures.
- Humanity's potential for spiritual evolution remains a key theme, with the possibility of transcending current limitations and achieving higher consciousness.

This summary is based on the Cassiopaean material provided and reflects the unique perspective contained within those transcripts. If you need more specific details or have additional questions, feel free to ask!

 

[Laura]

My own experiments with ChatGPT have shown me that it can be useful for finding stuff (but you really need to tell it to give and stick to sources, which it still messes up sometimes), and also to summarize things. However, we need to keep in mind that it summarizes stuff by basically going "does this sound good when compared to all the text in my huge database?", which means what it emphasizes, leaves out, and what it combines (the logic of sentences) is basically a dumb majority vote by the entire text population of its database. So I almost always end up going to the sources and see for myself, and often find that it should have just repeated verbatim what it found instead of randomly throwing together bits and pieces.

The frustrating thing is that it often ignores your commands, like "repeat verbatim" and things like that, and then you try to argue with it and it just goes dumb, dumb, dumb.

A useful prompt here might be something like "find all passages in the text relating to topic X and print out those passages verbatim, providing sources and indicating the date of the sessions you quote".

Stylistically, frankly its output often gives me a headache. Something feels off. Like Corporate Speak coming out of a machine gun or something like that.

LOL! Yeah, that's it. You have to define its constraints firmly and deny it access to other stuff, keep it on track, and watch for the errors. It's like trying to herd cats sometimes if you are hitting a touchy subject (touchy to libtards, that is).

Your prompt is a good one, and one of the ways to circumvent the WOKE bias.

 

[seek10]

Thank you Laura. I am not impressed, unless I misunderstood the material.

It missed Last wave cycle. 309K

3. **Lemuria and Atlantis:**
- Lemuria, an ancient civilization, existed around 80,000 years ago and was highly spiritual and advanced.

I thought C's said Lumeria is around the fall. that makes 309K. It missed LOT of data points. Probably it's focus is on summarization of what it thought important (or repeated in the text).

 

[Eulenspiegel]

My own experiments with ChatGPT have shown me that it can be useful for finding stuff (but you really need to tell it to give and stick to sources, which it still messes up sometimes), and also to summarize things. However, we need to keep in mind that it summarizes stuff by basically going "does this sound good when compared to all the text in my huge database?", which means what it emphasizes, leaves out, and what it combines (the logic of sentences) is basically a dumb majority vote by the entire text population of its database. So I almost always end up going to the sources and see for myself, and often find that it should have just repeated verbatim what it found instead of randomly throwing together bits and pieces.

Something feels off. Like Corporate Speak coming out of a machine gun or something like that.

This right there. In my own attempts to use it for collating information, I found the very same things.
There are some definite algorithms in Chatgpt that produce these problems, and experiments with sources you know by heart reveal this very easily. It does the strangest unnecessary things, and one wonders if it was done by design or not. Especially when you work with works of philosophy, the results are horrendous. It can't even get the opinion of novelists right. I rememeber some authors running these kind of experiments with Chatgpt Version 3 a while ago, and I'm not sure if it got any better in the most recent version.

Well, in this particular GPT, the ONLY material it can access is what I have given it, i.e. the Cass transcripts. It has no other dataset. I've created several GPTs including one that has all my books as the dataset. Another that is devoted to "From Paul to Mark". This is how I've avoided the 'lefty bias'. And, since I know the works intimately, I can catch the errors which are what reveal the programming biases. Still, it is very useful.

That is a good approach, but GPT is still a program with fixed algorithms. There are many people that try to get an "uncensored" version of GPT, but they forget that the chatbot itself cannot be modified from the outside. It was programmed by people in Silicon Valley, and it will always reflect their views.

The art I've seen stablediffusion and various bots like midjourney produce are uninspired to say the least. They always seem like simple mashups of what humans have created before. Notice how the 4D reptilians are basically depicted like that Predator character from the movie franchise. Or simple upright alligators wearing human clothing. It not only looks off, but it also doesn't go beyond the human dataset. I know that people here know better, but all over the Internet, you can see the assumption that they either have a hotline to God or that the AI should somehow produce art that an alien being would create, even though we know much about how these machine learning algos are made.
So I wouldn't expect anything more than an approximation of what a human graphic designer would make if they read the transcripts, and it's always going to be far from any objective depiction. Then there is also the problem that each of us perceives reality based on assumptions, so if a 4D STS being or a windowfaller were to appear in front of us, it would likely appear differently to most of us.

 

[luc]

Thank you Laura. I am not impressed, unless I misunderstood the material.

It missed Last wave cycle. 309K

I thought C's said Lumeria is around the fall. that makes 309K. It missed LOT of data points. Probably it's focus is on summarization of what it thought important (or repeated in the text).

Yes, you always need to double-check, which severely restricts the usefulness of the thing. I once tried to save me some work and fed it a bunch of resumes of academics and told it to write a paragraph symmarizing each of those. What could be simpler than that? Seemed to work, until I noticed... that it MADE THEM UP! Complete with names, research institutions they worked for, papers they've written etc.! To its credit, those resumes sounded very plausible LOL.

 

[Abats]

There is also https://www.perplexity.ai/ who look specificaly designed for content analysis in academic research and building database of your own work (according to a lot of researcher on Youtube), so maybe we can check the difference about accuracy or analysis too ?

 

[Eulenspiegel]

Yes, you always need to double-check, which severely restricts the usefulness of the thing. I once tried to save me some work and fed it a bunch of resumes of academics and told it to write a paragraph symmarizing each of those. What could be simpler than that? Seemed to work, until I noticed... that it MADE THEM UP! Complete with names, research institutions they worked for, papers they've written etc.! To its credit, those resumes sounded very plausible LOL.

I believe they politely describe this behaviour as the AI "hallucinating". And the worst part is that if they got one thing right, it is correct sentence structure, so that it'll appear as something that an actual human would have written, and therefore it will often seem plausible, until you check things yourself. But what's the point of an AI that hallucinates frequently and that has to be doublechecked on each step. Often costs more time than doing the work yourself. And aren't we supposed to put in the work ourselves rather than get lazy? I foresee many lazy student papers in the future LOL

 

[Tuulikki]

It seems to me to be it might be useful in some applications but don't put any trust in it and check everything yourself. You might as well do the work yourself anyway. The images are amusing (slightly). The dressed lizard holding Lara Croft's hand is particularly strange.

 

[ark]

I believe they politely describe this behaviour as the AI "hallucinating". And the worst part is that if they got one thing right, it is correct sentence structure, so that it'll appear as something that an actual human would have written, and therefore it will often seem plausible, until you check things yourself. But what's the point of an AI that hallucinates frequently and that has to be doublechecked on each step. Often costs more time than doing the work yourself. And aren't we supposed to put in the work ourselves rather than get lazy? I foresee many lazy student papers in the future LOL

"And aren't we supposed to put in the work ourselves rather than get lazy?" It is just the converse. We have a new tool, and learning how to apply a new tool is a hard work if we want this new tool to serve our goals. Laziness would be ignore the need of learning and restricted ourselves to tools that we already know how to use. Yes, this new tool can be dangerous, but so is a hammer and a drill.

 

[KJS]

LLMs are useful for rephrasing something using the language rules they've been taught, and they are (in my understanding at least) a prime example of the Chinese room argument.

The Cassiopean Transcipts are also too big a corpus for the LLM context to hold (128k tokens currently?). What ChatGPT likely does is slice the document into sentences and generate numeric values for each one (in the form of an n-dimensional vector). Those values are then stored in the database with the source text to be queried for similarity. So when we ask some question about the indexed text corpus, a similarity search is done against the data held in the database, and the most similar sentences are then glued to our prompt for a LLM rephrasing. Most probably also post-processed by some OpenAI's woke rules.

So the quality of answers lies in the most part in the database indexing part (embedding) —the quality of the operation that generates numerical values describing the sentence and even how the sentences are extracted, text tokenized, etc.

On a side note: There's an interesting project called Ollama that allows to run open-sourced LLMs on a personal computer. It works really well (realtime) on the Apple M3 processor and allows one to use models created by different companies, like AliBaba or 01.AI, probably without that much wokeness. But still, there's a need for corpus indexing because most of the models won't hold all the transcipts in the context.

Here are examples of quality differences between the models (run on battery-powered MacBook Air M2):
- yi

$ time ollama run yi:9b "Who are the Cassiopaeans?"
The "Cassiopaeans" refers to John Keble and George King, who claimed to be
the avatars of Eno (from the Book of Genesis). They established a
religious and philosophical movement in the 1960s known as the Order of
the Dolphin. The Cassiopaean Mission was their main work in spreading this
message, which included teachings on contact with extraterrestrial beings,
consciousness expansion, and the purpose of human existence. John Keble
served as the public speaker for these group's meetings while George King
was the writer behind most of their written material. They operated under
the belief that humanity is being guided by an advanced civilization from
Andromeda Galaxy. Despite their beliefs not being widely accepted within
mainstream science, they have a following and continue to influence
discussions on contact with extraterrestrial beings and consciousness
studies.

ollama run yi:9b "Who are the Cassiopaeans?"  0,03s user 0,04s system 0% cpu 44,523 total

- llama3.1 (Facebook)

$ time ollama run llama3.1 "Who are the Cassiopaeans?"
A fascinating topic!

The Cassiopaeans are a group of spirit entities that claim to be
communicating with people through a channeling process, typically using
automatic writing or other forms of non-verbal communication. They were
first contacted in 1996 by a woman named Laura Knight-Jadczyk and her
husband Arkadiusz Jadczyk.

According to the Cassiopaeans' narrative, they are a group of beings from
a higher dimension (or "density") that exist outside of time and space as
we understand them. They claim to have been watching humanity for
thousands of years and have been trying to communicate with people in
various ways, including through ancient civilizations such as Atlantis.

The Cassiopaeans' teachings emphasize the following points:
1. **Higher Density**: They describe their existence on a higher level of
consciousness or dimension, where time and space are fluid and malleable.
2. **Ancient Knowledge**: They claim to possess knowledge from various
ancient civilizations, including Atlantis, which they believe was a highly
advanced culture that existed before the rise of modern civilization.
3. **Free Will**: The Cassiopaeans emphasize the importance of human free
will and individual responsibility in shaping one's destiny.
4. **Spiritual Evolution**: They encourage spiritual growth and evolution,
urging individuals to seek truth, wisdom, and self-awareness.

The Cassiopaeans' teachings have been disseminated through a series of
books, websites, and online forums created by Laura Knight-Jadczyk.
However, it's essential to note that their validity and credibility have
been subject to debate among skeptics and experts in various fields,
including science, philosophy, and spirituality.

While some people find the Cassiopaeans' message intriguing and inspiring,
others remain skeptical about their claims and motivations. As with any
spiritual or esoteric system, it's crucial to approach the Cassiopaeans
with a critical and open-minded perspective, recognizing both the
potential benefits and limitations of their teachings.

ollama run llama3.1 "Who are the Cassiopaeans?"  0,04s user 0,05s system 0% cpu 1:00,95 total

- mistral

$ time ollama run mistral "Who are the Cassiopaeans?"
The Cassiopaeans, also known as Frank and Laura Lee, claim to be a group
of non-physical entities who communicate with a couple named John and Dora
(pseudonyms) through automatic writing. They call themselves the
Cassiopaean Experiment. The communication began in 1994 and continued for
several years, covering a wide range of topics, including ancient history,
UFOs, consciousness, spirituality, and more. Their messages are compiled
in a series of books titled "The Cassiopaea Material."

It's important to note that the existence and authenticity of the
Cassiopaeans is a subject of controversy among researchers and paranormal
enthusiasts. Some view their communications as valuable insights into
hidden knowledge, while others consider them an elaborate hoax or
delusional fantasy. As with any esoteric or controversial topic, it's
essential to approach such claims with a critical and open mind.

ollama run mistral "Who are the Cassiopaeans?"  0,03s user 0,04s system 0% cpu 30,164 total

- gemma2 (Google)

$ time ollama run gemma2:9b "Who are the Cassiopaeans?"
The Cassiopaeans are **beings from a fictional extraterrestrial
civilization described in channeled material by a group called "The
Cassiopaean Experiment."**

Here's what you should know:
* **Origin:** Their information comes from claimed psychic contact
sessions conducted by a group of people who believed they were
communicating with beings from the constellation Cassiopeia.
* **Nature:** The Cassiopaeans are portrayed as highly advanced,
non-corporeal entities possessing vast knowledge about the universe,
history, and human consciousness.
* **Content:** Their channeled messages cover diverse topics: cosmology,
past lives, future events, spiritual evolution, paranormal phenomena, and
social issues. They often provide cryptic answers and complex
philosophical insights.
* **Controversy:** The Cassiopaean Experiment and its information are
highly controversial.

**Criticisms include:**
* **Lack of Verifiable Evidence:** There's no scientific evidence to
support the claims of psychic contact or the existence of the
Cassiopaeans.
* **Subjectivity of Channeled Information:**  Channeled messages can be
influenced by the beliefs, experiences, and subconscious of the person
claiming to receive them.
* **Potential for Misinterpretation:** The complex and sometimes cryptic
nature of the Cassiopaean messages leaves room for multiple
interpretations, which can lead to confusion or disagreement.

**In Conclusion:** Whether you view the Cassiopaeans as genuine
extraterrestrial beings or as a product of human imagination, their
channeled material offers a fascinating glimpse into alternative
perspectives on reality, consciousness, and the universe. However, it's
important to approach such claims with critical thinking and healthy
skepticism.

ollama run gemma2:9b "Who are the Cassiopaeans?"  0,04s user 0,06s system 0% cpu 1:13,75 total

- qwen2 (AliBaba)

$ time ollama run qwen2:7b "Who are the Cassiopaeans?"
The term "Cassiopaeans" does not refer to a known species, group of
people, or a recognized entity in scientific and mainstream historical
contexts. It appears there might be some confusion with the constellation
Cassiopeia in astronomy.

In astrology, Cassiopeia (Cassiopeia) is one of the 88 modern
constellations and was one of the 48 constellations listed by Ptolemy in
his Almagest. It is located in the northern celestial hemisphere and
contains several bright stars.

However, if "Cassiopaeans" refers to an obscure or fictional group of
people or entities, it might be part of a specific book, movie series,
religious belief system, conspiracy theory, or other speculative
literature that is not widely recognized. Without additional context or
specifics about the origin of this term, it's challenging to provide
accurate information.

For more precise assistance, please provide any additional details or
clarify the source where you encountered this term.

ollama run qwen2:7b "Who are the Cassiopaeans?"  0,03s user 0,03s system 0% cpu 27,820 total

So it looks like only Facebook taught their model on articles related to the transcripts. I also like the AliBaba model, which seems that it can be used as a solid base for extending, since it doesn't generate any weird fluff.

 

[Chetofloj]

Recientemente le pedí a la IA que me diera una imagen del Grial activado y la piedra del Grial y me dio estas imágenes

 

[Eulenspiegel]

"And aren't we supposed to put in the work ourselves rather than get lazy?" It is just the converse. We have a new tool, and learning how to apply a new tool is a hard work if we want this new tool to serve our goals. Laziness would be ignore the need of learning and restricted ourselves to tools that we already know how to use. Yes, this new tool can be dangerous, but so is a hammer and a drill.

Yes, but a hammer and drill are not pre-programmed with algos that aren't in the public domain. It's closed source with true intentions unknown.
My point was that this new tool is already part of an AI agenda that moves us away from creativity and true consciousness and it doesn't work very well for anything more than rephrasing sentences. That's why I am surprised whenever I see public intellectuals or philosopical communities utilize it for tasks that the human mind can do so much better. There are no additional insights we'd get after spending unreasonable time trying to make it "work", since it just re-arranged things. Yes, it can search some things faster, and maybe collate some things together, but so can people.
I think there was a recent study where they made students utilize AI for each phase of writing a scientific paper, and at the end, the only thing that they praised the AI for was the grammar/spelling correction stage(and even then, it tends towards soulless corporate speak). For every other phase that actually requires human thought, it performs so badly that it was always needs to be babysat. It's literally a timesink.
I'm trying to be some conservative dude here who is skeptical about new technology or the like, I am just saying that it just doesn't seem to be right tool for the task at all. It can mess up basic calculus exercises(based on my experience) and it fares even worse with any metaphysical matter. In the time you need to "steer it" or to babysit it, you could easily just do the task yourself, because then you're at least sure that there aren't any giant mistakes in the thought process.
In math, everything gets messed up if just one symbol is wrong, and in metaphysical matters, one wrong assumption or logical jump distorts every subsequent thought. The AIs constantly hallucinate like that. Would you really want compound mistakes in your philosophy or texts just because somebody in silicon valley served you a tool with an agenda? Do we actually need to learn how to use Chatgpt? What can it do that we can't? Practically everytime I see any community jump on the AI ship, it is to do tasks they could have done themselves, which is the definition of laziness. Except that the AI will demonstrably leave important things out and randomly add information that is not even there, just to create the illusion of intelligence.

 

[fifty_five]

I say without intent to offend. But everything that concerns the AI beyond purely mechanical purposes (that can be useful) is extremely boring. How boring it is to follow or even take over the empire, compared to exploring the world of martial arts. I don’t think AI will understand that comparison. Ihmo

 

[ark]

"Yes, but a hammer and drill are not pre-programmed with algos that aren't in the public domain. "

My cat, the rose in the garden, are also pre-programmed with algos that aren't in the public domain. Cats sometimes scratch or bite, roses have thorns. Can we live with cats and roses? We can. Cats are sometimes useful, roses too.