17,740 Pages of Neidan Texts
44 books published in less than 4 ½ years (February 2022 - June 2026, or 52 months). Total number of pages: 17,740, equivalent to about 341 pages per month and about 11 pages per day. With few exceptions, each book contains “translations” of and “commentaries” to at least one Neidan text (or a text closely related to Neidan, Internal Alchemy). In a few cases, a large text is “translated” in a multi-volume book.
This is the output of Mr. Vitaly Filbert (I assume this is his real name). His books are published by “Daoist Cultivation Press” but at least some of them use the free ISBNs provided by KDP (a company belonging to Amazon).
The breakdown by year is as follows:
- 2022: 7 books, 1812 pages
- 2023: 6 books, 1346 pages
- 2024: 1 book, 172 pages
- 2025: 16 books, 5988 pages
- 2026: 14 books, 8422 pages (until end of June)
2025 was the most productive year, with 16 books for a total of 5,988 pages, equivalent to about 499 pages per month and about 16.5 pages per day.
But that is nothing compared to 2026. If the “translation” pace continues in the next 6 months (July-December) with the same rhythm as the first 6 months, it will result in 28 books for a total of 16,844 pages, equivalent to about 1,404 pages per month and about 46 pages per day.
I have never bought — and will never buy — any of those books. All data cited above are available on the Amazon pages for each of his books. (If you are curious, search “Vitaly Filbert” in your local Amazon store, and scroll down to the “Product details” for each book, making sure you select the paperback version.)
Clearly Mr. Filbert possesses super-human abilities. One can only admire (perhaps with a bit of jealousy) his amazing skills, only comparable to those of the best AI systems.
Now, I do not claim that Mr. Filbert is aided by some non-human entity in his prodigious output, because I have no evidence for that. But there is a point I'd like to mention.
Anyone who has experimented with AI translations of Chinese texts knows that AI is (usually) “intelligent” enough to ignore errors in the supplied text and produce a “correct” translation. And anyone who deals with Chinese digital texts knows that one of the most frequent errors (out of the myriad typically found in digital texts) is the replacement of the Chinese character 曰 with 日. (In virtually all cases, this is due to the OCR software used to digitize a text, which mistakenly reads 曰 as 日). The character 曰 (read “yue”) means “to say” or “to be called”, and the character 日 (read “ri”) means “sun” or “day”. The characters are very similar (曰 is just wider than 日), but the words are completely different.
The Amazon page for Mr. Filbert’s “translation” of the Daode jing or Book of the Way and Its Virtue (https://www.amazon.com/gp/product/B0BTV98KGV) contains a blurb longer than usual. In the blurb, Mr. Filbert discusses two Daode jing sentences from sec. 55, which he “translates” as “benefiting life is called luck/happiness” (益生曰祥) and “using the mind-heart to control Qi is called being strong and powerful” (心使氣曰強).
In 4 cases, Mr. Filbert’s text does not contain the correct character yue 曰 (“to be called”) but the wrong character ri 日 (“sun” or “day”). That is, the two sentences read 益生日祥 and 心使氣日強, respectively. Mr. Filbert even provides a mini-dictionary for those sentences, which includes these two identical entries:
- 日 - call; name
- 日 - call; name
One of several repositories of digital premodern Chinese texts that contains those Daode jing sentences with the wrong character (ri 日 instead of yue 曰) is the (often) disastrous “Chinese Text Project” (https://ctext.org/heshanggong/55/zh), which (usually) simply reproduces texts already available from other websites. In particular, section 55 contains these four sentences: 知和日常,知常日明,益生日祥,心使氣日強。(As you can see, it contains the 曰 / 日 error 4 times).
If you are interested in doing a quick experiment, go to your favorite AI system and copy-paste the following, using the 4 wrong characters: “Translate into English: 知和日常,知常日明,益生日祥,心使氣日強。” I tried, and as of today these are the results:
- ChatGPT: “To know harmony is called constancy; to know constancy is called illumination. To augment life is called an auspicious sign; when the mind directs the qi, it is called strength.
- Claude: “To know harmony is called constancy; to know constancy is called enlightenment; to augment life is called inauspiciousness; to let the mind direct the vital breath is called forcing.”
- Gemini: “To recognize harmony is to know the eternal law; to know the eternal law is enlightenment. To artificially strain/force life brings calamity; to allow the mind to control vital energy is forcefulness/aggression.”
(Brief technical digression: The most correct of those 3 translations is the one by Claude, but Mr. Filbert’s translation matches the one by ChatGPT — or maybe ChatGPT consulted Mr. Filbert’s book?)
It would be interesting, to continue the experiment, to “translate” a whole text — say, the Daode jing with its 81 sections — by prompting an AI system somewhat like this: “For each section, give me a translation and a mini-dictionary. Then add comments, precise but not over-technical, and generally understandable to a non-specialist reader interested in Daoist self-cultivation and especially Neidan. Be a bit mystical, but don’t exaggerate. At the end, add something on the possible relevance of that section in the life of a contemporary person. Each section, including original text, translation, dictionary, and comments should average between 1.5 and 2 printed pages in a 6 x 9 inches book”.
That would basically match the size, number of pages, and structure of Mr. Filbert’s “translation”. You could produce a book very similar to his book simply by copying a Chinese text from a digital repository and feeding it to an AI system with the prompt similar the above.
But I'm not suggesting you do that — there is already enough Neidan slop in books and social media. The actual purpose would be to find out whether an AI system is “intelligent” enough to match Mr. Filbert’s extraordinary skills.
If you are serious, the experiment shouldn’t end with the Daode jing. It should be repeated for all texts “translated” by Mr. Filbert in those 17,740 pages. Assuming a modest output of about 11 pages per day (which an AI system can do in a couple of minutes), how long would that require? About 4 ½ years.
Thanks to AI, you could produce a collection of Neidan texts as spectacular as the one produced by Mr. Filbert. You don't even need to know Chinese, or understand the difference between 曰 and 日. AI will take care of that.
Post Scriptum: Mr. Filbert’s series continues. On July 30, 2026, he published the 45th book in the series, made of 771 pages. Since the previous 44th book was published on June 24, 2026, he has “translated” 771 pages or about 21.4 pages per day in five weeks.

Comments
Post a Comment