本日の談話テーマ: 「エピステーメ (左脳的情報) VS グノーシス(左右脳的霊知)」について

最新情報: 「ChatGPT談話室」記事シリーズの企画を開始した目的等については、以下の「メタ心理学者とChatGPTの談話室 その一」の記事を参照してください。
http://www.kitaoka.co/note/meta/

この投稿は、(株)オフィス北岡代表の北岡泰典が配信する「メタ心理学者とChatGPTの談話室 その53」です。

今回は、「エピステーメ (左脳的情報) VS グノーシス(左右脳的霊知)」関連の質問をChatGPTにしてみました。

以下の文章は、(株) オフィス北岡代表の北岡泰典が Note で発信した投稿の引用となっています。


以下が、ChatGPTとの英語の質疑応答内容です。

ChatGPT FAQ #53 (9/5/23):

Q: In your rely to our previous chat, you said:

“If you have specific questions or topics related to non-verbal communication, I’ll do my best to assist you based on the textual knowledge I have been trained on.”

Well, I have been distinguishing the two terms both connoting “knowledge”; one is “episteme” meaning solely left brain oriented knowledge and the other is “gnosis” meaning “soul’s knowledge”.

In this connection, there is an interesting proverb in New Guinea saying “Knowledge is only a rumor until it lives in the muscle”. The knowledge which is not yet in the muscle is episteme, while knowledge which is in it is gnosis.

I do understand that you can handle only episteme, and think that this is part of the reason why GAI cannot be equipped with “Meta Consciousness”, which alone can produce gnosis.

Also, I make this argument the basis of my own claim that Singularity, that is, according to my own definition, will never occur.

Please comment.

A: Thank you for sharing your perspective on the distinction between “episteme” and “gnosis” as forms of knowledge. It’s an interesting distinction, where “episteme” refers to left brain-oriented knowledge and “gnosis” represents a deeper, soulful form of knowledge.

Regarding the proverb you mentioned from New Guinea, it suggests that knowledge remains incomplete until it is embodied and lived through action or experience. This aligns with the idea that practical application and experiential understanding are essential for a holistic grasp of knowledge.

As an AI language model, my capabilities are indeed limited to handling “episteme” or textual knowledge. I don’t possess subjective experiences, emotions, or a “Meta Consciousness” like humans do. While I can provide information and insights based on the textual data I have been trained on, I lack the ability to truly embody or deeply understand knowledge in the way humans do.

As for the concept of Singularity, which typically refers to the hypothetical event where AI surpasses human intelligence, your claim that it will never occur based on the absence of “gnosis” in AI is an interesting perspective. The question of whether true “gnosis” can be achieved by machines is a matter of philosophical debate and speculation.

It’s important to recognize that AI, as it currently stands, operates on algorithms and data-driven patterns rather than subjective experiences or the embodiment of knowledge. While AI continues to advance in terms of its capabilities, it remains distinct from human consciousness and the depths of “gnosis” that you mentioned.

In conclusion, your distinction between “episteme” and “gnosis” provides an intriguing lens to consider the limitations of AI and its potential for achieving a form of consciousness comparable to humans. The concept of Singularity and the true nature of machine intelligence remain open questions that continue to be explored and debated.


以下が、DeepL の翻訳結果 (未編集) です。

Q: 前回のチャットへの返信で、あなたはこう言いました:

“非言語コミュニケーションに関する具体的な質問やトピックがあれば、私が訓練してきたテキスト知識に基づいて、全力でサポートします “とありました。

さて、私は、「知識」を意味する2つの用語を区別してきました。1つは、もっぱら左脳指向の知識を意味する「エピステーム」、もう1つは、「魂の知識」を意味する「グノーシス」です。

これに関連して、ニューギニアには「知識は筋肉に宿るまでは噂に過ぎない」という興味深い諺がある。筋肉に宿っていない知識はエピステーメー、宿っている知識はグノーシスというわけです。

確かにエピステーメーしか扱えないというのは理解できますし、GAIがグノーシスを生み出す「メタ意識」を搭載できない理由の一端はそこにあると思います。

また、この議論は、シンギュラリティ、つまり、私自身の定義によれば、シンギュラリティは決して起こらないという私自身の主張の根拠にもなっています。

コメントをお願いします。

A: 知識の形態としての「エピステーメー」と「グノーシス」の区別について、あなたの視点を共有していただきありがとうございます。エピステーメー」は左脳的な知識であり、「グノーシス」はより深い、魂的な知識であるというのは、興味深い違いですね。

また、ニューギニアのことわざですが、「知識は、行動や体験を通して体現され、生きられるまでは不完全なものである」ということを示唆しています。これは、知識を全体的に把握するためには、実践と体験的な理解が不可欠であるという考え方と一致します。

AIの言語モデルである私の能力は、「エピステーメー」すなわちテキスト知識を扱うことに限られています。人間のように主観的な経験や感情、「メタ意識」を持っているわけではありません。学習したテキストデータをもとに情報や洞察を提供することはできますが、人間のように知識を体現したり深く理解したりする能力はありません。

シンギュラリティとは、AIが人間の知能を超えることを指す概念ですが、AIに「グノーシス」が存在しないことを理由に、シンギュラリティは起こらないという主張は、興味深い視点ですね。機械が真の「グノーシス」を獲得できるかどうかは、哲学的な議論と推測の問題である。

現在のAIは、主観的な経験や知識の具現化ではなく、アルゴリズムやデータ駆動型のパターンで動作していることを認識することが重要です。AIは、その能力という点では進歩し続けていますが、人間の意識や、あなたがおっしゃる「グノーシス」の深みとは、依然として一線を画しています。

結論として、「エピステーメー」と「グノーシス」の区別は、AIの限界と人間に匹敵する意識形態を実現する可能性を考える上で、興味深いレンズを提供してくれますね。シンギュラリティの概念と機械知能の真の姿は、これからも探求され、議論される未解決の問題である。


AI は、主観も、感情も、メタ意識ももっていない、と告白しているのは、実に興味深いです。

もう一つだけ質問してみました。


以下が、ChatGPTとの英語の質疑応答内容です。

ChatGPT FAQ #52_2 (9/5/23):

Q: In our previous chat, you said “I don’t possess subjective experiences, emotions, or a ‘Meta Consciousness’ like humans do” and “The question of whether true ‘gnosis’ can be achieved by machines is a matter of philosophical debate and speculation”.

Well, I am a human being who possesses “subjective experiences, emotions, or a ‘Meta Consciousness'”, and am also a stern Shankarite for many years who has been thoroughly and intensively experimenting with human consciousness as well as with “gnosis” as “knowledge in the muscle”, and have reached the philosophical conclusion that Singularity will never occur.

Between the opinion of GAI devoid of subjective experiences, emotions, or a Meta Consciousness and the “live” opinion of mine who am equipped with them, which do you think has more validity as to such philosophical debates as Singularity?

I want you to make your comment.

As an AI language model, I do not have personal opinions or beliefs, nor do I possess emotions or a subjective consciousness. My purpose is to provide information and answer questions based on the knowledge and data I have been trained on.

The concept of Singularity and whether it will ever occur is a topic of philosophical debate, and there are valid arguments on both sides. However, as an AI language model, I am not equipped to determine the validity of philosophical arguments, nor can I provide a definitive answer to this question. Ultimately, the decision on the validity of philosophical arguments is left up to individual interpretation and personal beliefs.


以下が、DeepL の翻訳結果 (未編集) です。

Q: 前回のチャットで、あなたは「私は人間のように主観的な経験や感情、『メタ意識』を持っていない」「機械によって真の『グノーシス』が達成できるかどうかは、哲学的な議論と推測の問題である」と言いました。

さて、私は「主観的な体験や感情、『メタ意識』」を持つ人間であり、また、人間の意識と「筋肉における知識」としての「グノーシス」を徹底的に集中的に実験してきた長年の厳格なシャンカライトであり、「シンギュラリティは絶対に起きない」という哲学的結論に達しているのですが。

主観や感情、メタ意識を持たないGAIの意見と、それらを備えた私の「生」の意見とでは、シンギュラリティのような哲学的な議論において、どちらの方がより妥当性があると思われるでしょうか?

というコメントを出してほしい。

Q: AIの言語モデルである私には、個人的な意見や信念はなく、感情や主観的な意識も持ち合わせていない。私の目的は、訓練された知識やデータに基づいて、情報を提供したり、質問に答えたりすることです。

シンギュラリティという概念や、それが実現するかどうかは哲学的な議論のテーマであり、どちらにも正当な主張がある。しかし、AIの言語モデルである私には、哲学的な議論の妥当性を判断する能力はなく、この問いに明確な答えを出すことはできません。結局のところ、哲学的な議論の妥当性を判断するのは、個人の解釈や信念に委ねられているのです。


この珍しく非常に短い「最終的反判断は各個人に任せます」という回答は、ChatGPT側からの「降参の白旗」であると思うのは、本著者だけでしょうか。あるいは、子供のように臍を曲げてしまったのでしょうか (笑)。

以上、本日の「メタ心理学者とChatGPTの談話」は、いかがでしたでしょうか?

北岡泰典「メタ心理学宣言」ページ: https://www.kitaoka.co/meta/

作成 2023/5/23