|
 |
|
|
|
|
|
|
「人工智能 —人形機器人」能否被納入 佛教倫理世界的六道輪迴裡？=Could AI-Humanoid Robots be Included within the Six Paths of Cyclic Existence (Saṃsāra) of the World of Buddhist Ethics ? |
|
|
|
| 著者 |
蔡瑞霖 (著)
|
|---|
| 掲載誌 |
玄奘佛學研究=Hsuan Chuang Journal of Buddhism Studies
|
|---|
| 巻号 | n.44 |
|---|
| 出版年月日 | 2025.09 |
|---|
| ページ | 73 - 106 |
|---|
| 出版者 | 玄奘大學 |
|---|
| 出版サイト |
http://ird.hcu.edu.tw/front/bin/home.phtml
|
|---|
| 出版地 | 新竹市, 臺灣 [Hsinchu shih, Taiwan] |
|---|
| 資料の種類 | 期刊論文=Journal Article |
|---|
| 言語 | 中文=Chinese |
|---|
| キーワード | 人工智能（AI）=Artificial Intelligence (AI); 人形機器人=Humanoid Robot; 六道輪迴=Six Paths of Cyclic Existence (Saṃsāra); 倫理學=Ethics; Asura |
|---|
| 抄録 | 人工智能（ Artificial Intelligence, AI)）作為高科技工具，係建立在大型語言模式（ LLM）上，以運算法（ algorithm）處理超大量數據，模仿人類思維能力，以學習、推理及自我更正等認知能力為主發展出來的。最近，科技巨頭之一OpenAI推出ChatGPT和GPT4聊天系統，強調文字圖像、影像、聲音之驚人的生成能力，經過媒體大肆報導，網路使用者趨之若鶩，議論不斷——認為AI功能強大又危險，必須警告，要求延後發展，要訂定AI憲法與配套規範。加上工業、軍事、商業及家用各種型態機器人的動力機械技術不斷突進，透過雲端運算、大數據運用和遠距控制，可以被大量部署，來與AI更緊密的結合，其中人形機器人 humanoid robot）的發展，尤其矚目。也就是說，從模仿到深度學習，從博弈學習、隨機決策到自主數據驅動，AI都將跟緊著人類足跡而無處不在。
不論AI-robot或AI humanoid being，更受關注的人工通用智能 AGI），甚至是超智智能（ super intelligence）都影響著人類的未來。那麼，它（ 她／他）到底有無意識、有無情緒、有無自主決定與行為能力的問題，實質上已經觸及倫理學問題，可能引發各種道德難題。目前我們還無法在定義、分類、監察與究責上，把這類AI視為明確的行為者或代理人，所以不論教育學習互動、醫療照顧、軍事衝突、社會安全各方面，既有受益期待也都藏有隱憂。如何定位這類存有，如何正確發展AI，確實是重要和迫切的問題。
本文嘗試從佛教緣起觀來看上述問題，尤其借用阿修羅（ Asura）秉性和「非人非天」特徵，來類比於當代AI（、robot（、和AI humanoids這些存有類別，並且分析、預測其個體、成立的依據、行為獨立性等，進一步反思有情眾生在當代資訊時代的業力處境，以及如何解開繫縛的問題。爰此，本文觸及的子項目將包括：
（1）人工智能、仿生的人形機器人如何定義？是否它將（ 正在）發展為獨立自主行為的個體？如既有「機器人三定律」宣言或未來《機器人法》立法等，可否視為它具有自身的業力？
（2）人工智能、雲端運算、大數據及遠端遙控，如何以自我學習方式介入到人類生活中，而產生共構與共業行為？對佛教六道輪迴的一體完備性的說明有何影響？
（3）既有的佛教倫理學，能否有效應用到AI的發展的新課題上？傳統的律宗及戒學內容有可以援用來進行規範之參考的嗎？等等，亦將引用一些統計數據、實例及實證來論述之。
As a high-tech tool, Artificial Intelligence (AI) is built on the Large Language Model (LLM), processes a large amount of data with algorithms, imitates human ability of thinking, and develops its “learning, reasoning, and self-correction” and other cognitive capacities. Recently, Open AI, one of the technology giants, launched the ChatGPT chat system, emphasizing the amazing power to generate text, images, video, and sound. After extensive media reports, Internet users flocked to it, and there were constant discussions — AI is powerful and dangerous, so we must warn, it is required to temporarily postpone the development and formulate an AI constitution and supporting norms in regulation. Coupled with the continuous advancement of power machinery technology of various types of robots in industry, military, business, and household use, through cloud computing, big data application, and remote control, they can be deployed in large numbers to be more closely integrated with AI. Among them, is the development of the humanoid robot, especially attention. That is to say, from imitation to deep learning, from game learning, random decision-making to autonomous data-driven, AI will follow human footprints and be everywhere.
Regardless of AI-robot or AI humanoid being, artificial general intelligence (AGI), which has received more focus, and even super intelligence will affect the future of mankind’s life. Then, the question of whether it (she/he) has consciousness, emotion, independent decision-making, and behavioral responsibility has touched on ethical issues in essence and may cause various moral problems. We are still unable to regard this type of AI as a clear actor or agent in terms of definition, classification, monitoring, and accountability. Therefore, regardless of education and learning interaction, medical care, military conflict, and social security, both benefits are expected and hidden worries. How to locate such beings and develop AI correctly are important and urgent issues.
This article attempts to look at the above-mentioned issues from the perspective of Buddhist interdependence, especially borrowing Asura’s disposition and its “non-human, non-heaven” characteristics to compare it to contemporary AI, robot, and AI humanoids, and to analyze and predict their individual, the basis for its establishment, and the independence of behavior, etc., to further reflect on the karmic situation of sentient beings in the current information age and how to untie the bondage. Therefore, the sub-items covered in this article will include:
(1) How to define artificial intelligence and bionic humanoid robots? Is it developing into a self-acting individual? For example, the declaration of the “Three Laws of Robotics” or the legislation of the “Robot Law” in the future, and that can be regarded as having its particular karma？
(2) How can artificial intelligence |
|---|
| 目次 | 一、AI世俗語言是大共業造成的 77
（一）語業生成一切 77
（二）誰是AI Asura ? 80
（三）提示詞：繞著 Asura 聊 AI 82
（四）對話即生成文本 84
（五）AI樣態總是本真的嗎? 86
（六）AI的具身化問題 89
二、從佛法也是戲論來看 AI 倫理規範 90
（一）AI的自我確認:只存在運算判斷中 91
（二）只要是行為個體或代理人,都可被追責 91
（三）AI 規範與律則之虛構與否 92
（四）AI決策的倫理依據 93
（五）AI Asura 的對齊問題 94
三、AI模仿、模擬到擬仿是性空唯名嗎? 95
（一）科技業界不喜歡倫理與道德審查委員 95
（二）業力與發願 97
（三）關於超級AI的發展:臨近奇點 98
（四）「數據糾纏」現象與「緣起性空」的體驗 100
四、簡短結語 101
參考書目 104 |
|---|
| ISSN | 18133649 (P) |
|---|
| ヒット数 | 1 |
|---|
| 作成日 | 2026.01.20 |
|---|
| 更新日期 | 2026.01.20 |
|---|
 |
Chrome, Firefox, Safari(Mac)での検索をお勧めします。ＩＥではこの検索システムを表示できません。
|
|
