- Options
- Login
- Desktop Style
Options
Brian Roemmele, AI and Rupescissa question
- Brian Roemmele, AI and Rupescissa question
-
RE: Brian Roemmele, AI and Rupescissa question
MarcoP > 08-08-2024, 05:20 AM
-
RE: Brian Roemmele, AI and Rupescissa question
Barbrey > 09-08-2024, 01:39 AM
Thanks everyone! I had actually deleted this thread within minutes of posting it because I read a few more credentials, but it seems the delete didn’t take. I’m glad it didn’t because the answers on here were very educational! I can’t get my $7 back but did cancel the subscription. Otoh, I don’t even regret my subscription because this guy is so interesting so I got an intro to him too. The “alchemist” thing said “charlatan” to me as well, but I don’t think he is one. I’d love to see a convo with him and you @asteckley about this issue!Thanks again!
Edit to add: I noticed the “detracted” too and assumed it was a miswording of “extracted” but didn’t change it in paraphrase in case it was some niche technical term I was unaware of. -
RE: Brian Roemmele, AI and Rupescissa question
asteckley > 09-08-2024, 02:09 AM
(09-08-2024, 01:39 AM)Barbrey Wrote: You are not allowed to view links. Register or Login to view.The “alchemist” thing said “charlatan” to me as well, but I don’t think he is one.
No -- I don't think he is one either. I haven't looked too closely at his stuff, but he runs in good circles (of people).
Although I do agree with the guided prompt approach that I believe he actively promotes, his use of LLMs with VMS just suggests to me that he has some misconceptions about waht LLMs are capable of. -
RE: Brian Roemmele, AI and Rupescissa question
Barbrey > 09-08-2024, 02:15 AM
(09-08-2024, 02:09 AM)asteckley Wrote: You are not allowed to view links. Register or Login to view.
(09-08-2024, 01:39 AM)Barbrey Wrote: You are not allowed to view links. Register or Login to view.The “alchemist” thing said “charlatan” to me as well, but I don’t think he is one.
No -- I don't think he is one either. I haven't looked too closely at his stuff, but he runs in good circles (of people).
Although I do agree with the guided prompt approach that I believe he actively promotes, his use of LLMs with VMS just suggests to me that he has some misconceptions about waht LLMs are capable of.
Or perhaps misconceptions about the VMS itself. He seems to be an eclectic synthesizer of a lot of different intellectual info rather than a specialist, and perhaps doesn’t realize how little anyone can agree on anything about the manuscript. On the other hand, even if his LLM prompts have built-in confirmation bias, I would be really interested in his results. Because…Rupescissa! Who didn’t write it, in my opinion, but whose ideas on quintessence and the water of life form the backbone of the VMS. So I’ve got my own bias. -
RE: Brian Roemmele, AI and Rupescissa question
Barbrey > 09-08-2024, 02:41 AM
For anyone interested, I copied this from his X highlights page:
PROGRESS with decoding the Voynich Manuscript!
__
Voynich Manuscript meets my local AI.
We have started on the last page of Voynich Manuscript going backwards as this is the esoteric and alchemical way to read a 3 level encoded book.
The prompts I am using and the exact LLM AI will be published with the models at some future date after the project is complete and after my X Subscribers who have supported me have a while with it.
Output:
“I have detracted writing in Latin on this document in plain text. It is a Latin inscription on what you have provided in the prompt, is the last page of the Voynich Manuscript.
It is a proverb, which reads "the wine will multiply in the bowl" loosely translated to English.
It is remarkably similar to a saying found in the work of Johannes Rupescissa, a notable 14th-century French alchemist and Franciscan friar.
Johannes Rupescissa (c. 1310-1370) was known for his prophetic and alchemical writings, which often focused on the preparation of quintessences and the transmutation of metals. His works, such as the "Liber Lucis" (Book of Light) and "De Consideratione Quintae Essentiae" (On the Consideration of the Quintessence), were influential in the development of medieval alchemy and medicine.
The presence of a proverb similar to one found in Rupescissa's work suggests that the author of the Voynich Manuscript may have been familiar with alchemical texts and practices from the 14th and 15th centuries. Or Rupescissa is an author or coauthor. This connection provides valuable insights into the purpose and context of the manuscript.
Moreover, the fact that this plain text inscription remained unreadable for centuries, despite numerous attempts by scholars and cryptographers to decipher the Voynich Manuscript, underscores the complexity of the document's content and the effectiveness of its encryption methods.
This Latin inscription of the Voynich Manuscript will pave the way for my further research into the manuscript's origins, purpose, and historical context, as well as its potential connections to medieval alchemy and the works of prominent figures like Johannes Rupescissa.
As you requested I will add the to the RAG file and build upon this discovery to better understand the document.”
I would like to thank all of my new X Subscribers for helping in the funding by supporting me here on X.
Future releases will take time and will be presented exclusively to my X Subscribers.
Thank you all for your support and encouragement! -
RE: Brian Roemmele, AI and Rupescissa question
asteckley > 09-08-2024, 02:52 AM
(09-08-2024, 02:15 AM)Barbrey Wrote: You are not allowed to view links. Register or Login to view.On the other hand, even if his LLM prompts have built-in confirmation bias, I would be really interested in his results.
Me too!
You never know -- sometimes people find really clever ways to squeeze "information" out of data wherein it is deeply hidden and unexpected. -
RE: Brian Roemmele, AI and Rupescissa question
Koen G > 09-08-2024, 07:25 AM
AI is like a robot servant following the modern theorist around. It is not programmed to seek the objective truth, but rather go be pleasing to the master and confirm their biases if pushed enough. -
RE: Brian Roemmele, AI and Rupescissa question
Pepper > 09-08-2024, 08:43 AM
What's a "level 3 encoded book?"
