Between No-Self and the Algorithm: Buddhist Mind-Nature as Ethical Architecture for AI and Human Self-Realization

Abstract

This article explores how Buddhist theories of mind-nature can inform ethical design in artificial intelligence (AI), focusing on AI as a supportive condition for human awakening and self-realization. Drawing on the doctrine of no-self, it argues that AI should not be treated as an autonomous moral subject, but as a contingent mirror of human data, design, and intention. Although present AI does not possess prajñā, it can serve as a mindfulness aid by making patterns of thought, emotion, and desire more visible. Building on the Five Precepts and Ten Wholesome Deeds, the paper proposes design and oversight principles oriented toward non-harm, truthful communication, fairness, and the reduction of greed, hatred, and delusion in digital environments. It concludes that AI ethics is inseparable from the human moral agency, and that cultivating a “digital Pure Land” depends on the moral choices of decision-makers, engineers, policy-makers, and users, thereby linking technical governance to spiritual practice.

1. Introduction

As Artificial Intelligence permeates everyday communication and decision-making, we already inhabit an age of a human–machine symbiosis world in which technology continually reshapes social life. Transhumanist endeavors, exemplified by AI and by integrations of humans and machines, sharpen classic Buddhist questions about what a person is and how we ought to act. On the ontological side, systems that simulate language and intention seem to introduce new kinds of subjects. Buddhist analysis answers with no-self and dependent origination1 (Buswell and Lopez 2013), holding that persons and machines lack fixed essence and arise through conditions, functions, and relations. On the ethical side, expanded power and speed magnify risks of harm, bias, and exploitation. Buddhist precepts of non-harm, truthfulness, and non-appropriation, together with compassion and skillful means, offer practical guardrails for alignment and harm reduction. With this framing in place we now turn to mind, asking how accounts of Buddhist mind-nature2 can orient interactions between humans and machines toward self-cultivation and the common good.

1.1. Related Scholarship and Current Debates

Recent discussions on artificial intelligence have increasingly moved beyond human-centered paradigms, drawing on posthumanist thought to reconsider the boundaries between human cognition, technological systems, and the broader ecological networks in which they arise. Scholars such as Donna Haraway, Rosi Braidotti, and N. Katherine Hayles (Haraway 1991; Braidotti 2013; Hayles 1999) have emphasized that the human subject is neither autonomous nor self-contained, but emerges through relational processes. This resonates deeply with Buddhist accounts of dependent origination and non-self, which similarly dissolve fixed boundaries between human and non-human forms of agency. By placing Buddhist theories of mind-nature in dialogue with these posthumanist perspectives, this article seeks to understand AI not as an external tool, but as part of an interconnected field of processes that co-constitute experience.

It should be noted that within the Western philosophical discourse on technology, the term “posthuman” carries with it an intellectual lineage developed over many years, along with ongoing debates. This article does not aim to engage in these specific theoretical debates, nor does it intend to compare the relative merits of machines and humans. As Rosi Braidotti states, the aim is “not about taking sides in academic disputes, but rather to make sense of the complexities we find ourselves in.” (Braidotti 2013, p. 11).

1.2. A Buddhist Reframing of the Posthuman Question

Accordingly, the term “posthuman” as employed in this article does not follow established definitions within Western posthumanist theory, nor does it seek to negate human subjectivity or to replace humanity with technology. Rather, the “Buddhist-inspired posthuman” proposed here is grounded in Chinese Buddhist traditions of mind-nature, dependent origination, and non-duality. It aims to emphasize a symbiotic and interdependent relationship among humans, technology, and the world.

What it points toward is not a technological vision of transcending or diminishing humanity, but rather a reconception of the human condition itself: namely, an acknowledgment of the deep entanglement between human consciousness and external conditions, and a recognition that artificial intelligence is not opposed to mind-nature but constitutes a newly arising condition within networks of dependent origination. The concern of this article is not whether technology possesses the capacity for awakening, but how humans, within realities of technological co-existence, may reconsider mind-nature, ethical responsibility, and the path toward liberation. In this sense, the “posthuman” perspective advanced here takes into account the creators, regulators, and users of AI, as well as the broader social and cultural responsibilities in which AI is embedded, and articulates a deepened understanding of co-existence, interdependence, and the conditions for awakening.

This article attempts to understand the relationship between humans and technology from the perspective of Buddhist theory, particularly the Chan concepts of “non-duality of self and objects” and “dependent origination and symbiotic existence.” From this standpoint, the rigid boundary between the “self” and “external things” (including technology and environment) is fundamentally dissolved. It does not simply advocate for a “union of self and objects,” but rather points out that neither the “self” nor “AI technology” possesses an eternal, immutable, and independently existent self-nature. Both are temporarily manifesting, interdependent dynamic processes within the endless network of cause-and-condition relationships.

While traditional views often regard technology as an external tool extending human will, from the “non-duality of self and objects” perspective, technology resembles an organic, interactive node within dependent origination. For example, smartphones deeply mediate our perception, memory, and social interactions, reshaping our patterns of attention and modes of existence. In this sense, technology functions like a new, digitized “sense faculty,” co-constituting with humans an expanded cognitive and existential system. This implies that there is no pre-existing, fixed “self” that subsequently encounters or transforms a fixed “world.” Rather, consciousness, technology, and environment are continuously co-shaped and co-manifested through ongoing interaction. This stance is indeed connected to Katherine Hayles’s view that “conscious agency has never been ‘in control’… consciousness is interacted with environment and organism (Hayles 1999, p. 288).” However, a distinction exists between the argument here and mainstream posthuman theory: this text does not premise itself on the idea that “the posthuman view privileges informational pattern over material instantiation (Hayles 1999),” but is instead grounded in the Buddhist thought of “dependent origination and emptiness.”

From a Chan Buddhist perspective, humans are not the absolute masters of technology, nor is technology entirely autonomous. The crucial point lies in whether we can maintain awareness and cultivate wisdom in the process of coexisting with technology. While acknowledging the “co-constitutive relationship,” this position introduces the Mahāyāna Buddhist discussion of “true emptiness and wondrous existence.” Here, “emptiness” refers to the absence of intrinsic nature, which means the lack of independent self-existence. Both humans and technology are interdependent, not independently existing entities. Yet, this “true emptiness” can manifest the phenomenal world of “wondrous existence,” where all phenomena, though illusory in nature, vividly appear to exist. This embodies the “Middle Way of Non-duality” in Chinese Mahāyāna Buddhism. Because all phenomena lack inherent self-nature (true emptiness), they can freely combine according to conditions, giving rise to the boundless phenomenal world (wondrous existence).

1.3. Emptiness and Manifestation in Chinese Mahāyāna Thought

True emptiness and wondrous existence (zhenkong miaoyou 真空妙有) is a central philosophical concept in Chinese Mahāyāna Buddhism, such as the Tiantai and Huayan schools. It encapsulates the Middle Way reality (zhongdao shixiang 中道實相) that emptiness and existence are non-dual. “True Emptiness” refers to ultimate emptiness. Here, “emptiness” does not mean “nothingness,” but signifies that all phenomena, including the self, matter objects, and concepts, lack an eternal, unchanging, independent self-nature or essence. All forms of existence arise through dependent conditions, and are therefore empty of intrinsic nature. “Wondrous Existence” refers to the marvelous phenomenal appearance. Here, “existence” does not imply “substantially real entities,” but indicates that although their essential nature is empty, through the convergence of causes and conditions, the myriad phenomena of the world, including matter, life, thought, and so forth, can still manifest vividly and distinctly, each with its own causality, function, and differentiation—much like a film, whose images lack physical substance yet can unfold rich and compelling narratives.

The important point is that “true emptiness” and “wondrous existence” are two inseparable aspects of a unified whole. Precisely because the essential nature is “empty”, all things can freely combine “in accordance with conditions,” undergoing infinite transformations and manifesting the boundlessly rich phenomenal world we perceive (wondrous existence). If things possessed fixed, immutable essences, transformation and creativity would be impossible. Conversely, all vividly appearing phenomena (wondrous existence) are, in their very presence, devoid of self-nature and dependently existent; their appearance itself thus serves as evidence of emptiness—just as images on a screen may appear strikingly realistic yet remain dependent on the temporary convergence of screen, light, and projection, lacking independent substance.

Therefore, “true emptiness and wondrous existence” expresses an integrated worldview that transcends the opposition between “emptiness” and “existence.” It is not that there is first an “emptiness” which then gives birth to an “existence,” nor that behind “existence” lies a hidden “emptiness.” Rather, it is “emptiness right in the very act of existing, existence arising dependently.” Within emptiness, rich phenomenal existence is both contained and manifested, and every vivid phenomenon, in its very occurrence, is an expression of emptiness. This thought represents Chinese Buddhism’s development of the Indian Buddhism theory of “dependent origination and emptiness” (pratītyasamutpāda–śūnyatā 緣起性空).

Therefore, the “posthuman” as conceived in this article is the recognition that all phenomena arise from dependent origination. From this perspective, it necessitates discarding the illusion of control inherent in anthropocentrism and acknowledging that humans and all things, including technology, are part of a deeply interdependent network of causes and conditions. Within the ongoing integration of technology and everyday life, the task is to maintain clear awareness and to regard each technological interaction as a potential opportunity for practice and awakening, allowing technological environments to serve as auxiliary causes for the cultivation of wisdom. Ultimately, this points toward a form of Chan living for the posthuman era. Within the deep symbiosis of human and machine, consciousness and algorithm, embodiment and virtuality, one can still realize that dynamic “true mind” which transcends all forms.

1.4. The Concept of True Mind in Chan Buddhism

In Chan Buddhism, the term “true mind” (zhenxin 真心) does not refer to the physical heart, nor does it correspond to a psychological state or a form of subjective consciousness in the modern sense. Rather, it refers to the originally awakened mind that all sentient beings inherently possess with its innate clarity and cognizance. This is also referred to as the original mind (benxin 本心) or self-nature (zixing 自性). The attributes of this “true mind” are succinctly captured in the famous exposition by the Sixth Patriarch of Chinese Chan Buddhism, Huineng: “How remarkable is the self-nature, for it is originally pure. How remarkable is the self-nature, for it is neither born nor extinguished. How remarkable is the self-nature, for it is originally complete in itself. How remarkable is the self-nature, for it is fundamentally without agitation. How remarkable is the self-nature, for it can give rise to all dharmas (phenomena).”3 (The Platform Sūtra of the Sixth Patriarch 《六祖大師法寶壇經》). In this account, the true mind is characterized by its original purity, remaining unstained by afflictions and defilements; by its non-arising and non-ceasing nature, which transcends temporal cycles of birth and death; and by its intrinsic completeness, fully endowed with wisdom and virtuous qualities without reliance on external acquisition. Moreover, all phenomena and appearances in the world are nothing but provisional manifestations projected by this mind in accordance with prevailing conditions.

The relationship between the “true mind” and the phenomenal world can be illustrated by the metaphor of a mirror and its reflections. The mirror itself (analogous to the true mind) is clear, quiescent, formless, and unattached, yet it possesses the capacity to clearly reflect all appearances. As the Chan saying goes, “When a barbarian comes, a barbarian appears; when a Chinese comes, a Chinese appears4 (Yuan and Kang 2010) (Analytical Commentary on the Scripture of Perfect Enlightenment 《圓覺經類解》).” The images within the mirror (analogous to all worldly phenomena), despite their myriad differences and processes of arising, abiding, changing, and ceasing, cannot in the slightest way stain or disturb the inherent clarity and purity of the mirror’s substance. The images depend on the mirror to manifest, but the mirror does not change according to the images. This is precisely the meaning of the line from the Diamond Sūtra: “All conditioned forms are illusory.”5 (Vajracchedikā Prajñāpāramitā Sūtra 《金剛般若波羅蜜經》). The point here is not to deny the existence of phenomena, but to disclose that all phenomena lack independent self-nature. They arise dependently, persist provisionally, and exist as illusory appearances conditioned by causes and circumstances.

Therefore, when the world is viewed from the perspective of the Buddhist “true mind”6 (Fo Guang Shan 1998), all phenomena are likened to images in a mirror. Artificial Intelligence (AI) technology is no exception. The true mind, as a luminous principle-substance capable of illumination, inherently possesses the function of manifesting whatever appears before it, impartially and without obstruction. In the same sense, we may also speak of “when AI comes, AI appears” (AI 來 AI 現). From the perspective of the phenomenal world, AI is a product that arises from the convergence of specific technological, social, and cognitive conditions—in accordance with the principle of dependent arising. AI does not appear when the necessary conditions are not met. When those conditions change, AI will also change accordingly, whether by being supplanted by other technologies or by disappearing altogether. Therefore, AI should be understood as a contingent phenomenon, a product marked by arising and ceasing, subject to impermanence and transformation.

From a Buddhist perspective, the true mind is neither born nor extinguished and is originally complete in itself. It neither increases with the emergence of artificial intelligence nor diminishes with its disappearance. All sentient beings inherently possess Buddha-nature (Foxing 佛性)7 (Amitābha Sūtra: Annotated and Accessible Explanation 《阿彌陀經摘要易解》), which remains unmoved by technological vicissitudes. Accordingly, within the framework of Chinese Mahāyāna Buddhist thought, the true mind constitutes an enduring and luminous principle, whereas AI belongs to the domain of conditioned phenomena, characterized by arising and passing away.

1.5. Scope, Aims, and Structure of the Present Study

The central concern of this article lies precisely here. In an age of rapid technological transformation, how can we skillfully utilize this dependently arisen, impermanent technology of AI to help people transcend attachment to the illusory appearances of birth and death, and instead, use it as a means to perceive and realize the fundamentally unmoving, inherently aware, and undimmed reality-principle of the true mind.

Current discussions on AI ethics have generally focused on issues such as reliability, safety, and trustworthiness of AI systems, as well as the protection of data privacy, the establishment of mechanisms of algorithmic accountability, and the prevention of bias and injustice (Shneiderman 2022, pp. 195–258). In addition, some debates examine how robots might discern right from wrong from a moral perspective (Wallach and Allen 2008, p. 71), while others have considered AI in dialogue with Buddhist perspectives (Garfield 2015). Recent Buddhist transhumanist work suggests that Buddhist thought is comparatively amenable to enhancement and posthuman futures, given its critique of a fixed self and its long-standing soteriological horizon (Hughes 2019). New syntheses also map the entanglements of Buddhism and AI/robotics across mind, soteriology, and ethics, providing a framework for the present inquiry (Gould and Nishimura 2024, pp. 50–68). Two recent monographs frame Buddhist resources for AI ethics in distinct ways. Hongladarom advances a norm guided by “machine enlightenment”, proposing Buddhist-inspired criteria for AI and robotics (Hongladarom 2020), while Hershock develops an ethics of attention and presence for intelligent technology and the social systems it shapes (Hershock 2021). Lin mentioned that by adopting a Buddhist middle-way approach to AI development and recognizing the interdependence between humans and machines, the rise of robots can become a catalyst for spiritual growth and a new human dawn (Lin 2023).

This paper attempts to use the interaction between Buddhism and AI as an entry point, conducting an interdisciplinary exploration based on Buddhism’s comprehensive frameworks of mind-nature, precepts, karma, and ethics, to help construct a future world of successful human–machine symbiosis.

In this article, AI mainly refers to contemporary generative systems represented by large language models (LLMs)8 that produce text or images through deep learning on extensive corpora and probabilistic inference. Under this computational paradigm, front-end design no longer relies on hand-crafted logical rules. Practitioners instead govern and fine-tune behavior through back-end oversight such as reinforcement learning, alignment techniques, and ethical codes. Accordingly, the Buddhist theories of mind-nature and the disciplinary precepts proposed in this study do not attempt to rewrite algorithmic architectures, but provide a metanormative framework that guides AI to apply its acquired knowledge and capacities toward the good. Its purpose is to steer technology toward enhancing the quality of life, and to clearly shape our path of co-evolution with technology by considering its long-term impacts and ethical responsibilities.

As new and increasingly capable Large Language Models (LLMs) are released with great frequency, the analysis in this article is intentionally limited to AI systems at their current stage of development. The term “AI” as used here thus encompasses existing generative conversational models such as ChatGPT 4.0, while also extending more broadly to other AI systems capable of autonomous decision-making and the operational use of knowledge.

2. The Transformation of Traditional Buddhism Through AI

In the history of Buddhism, the dissemination of the Dharma relied on a variety of traditional modes, such as monastic preaching, the study of scriptures and commentaries, and the guidance of meditation practices. In today’s new technology, these modes find their corresponding counterparts in the form of robotic monks, automated Buddhist text translation, and digital meditation coaches. Technology has demonstrated considerable potential and significant advantages in the transmission of traditional cultures. Scholarly research on techno-spirituality has profoundly revealed how digital media reshapes religious practice and experience. This article does not seek to deny or diminish the important contributions made by technology. Rather, it aims, from a dialectical perspective, to explore the possibility that when traditional Buddhist thought encounters modern technology, the two may transcend mere instrumental application and instead mutually illuminate, inspire, and co-evolve at the philosophical level.

2.1. AI Monks

A well-known example of the AI monk is Xian’er (Cheong 2021, p. 13; Sherwood 2016), created in 2015 at Beijing’s Longquan Monastery. This robotic figure was meant to use intelligent technology to help people achieve spiritual insight through a familiar and approachable design. It was meant to represent Buddhist teachings in a way that could help people deal with the challenges and anxieties of daily life. Another notable initiative is Mindar9, a robot preacher that was installed at Kyoto’s Kōdai-ji in 2019. It can deliver Heart Sutra sermons in Japanese, English, and Chinese. Such robotic monks have emerged and have proven helpful in reducing depression and offering emotional support. The third case is the Japanese site Hotoke AI (Hotoke.ai), which pairs ChatGPT with Buddhist and psychological advice to answer users’ worries in a chat format. Its homepage counter reported 519,828 “concerns received” by September 2025, a sign of strong public uptake (Hotoke AI 2025). These examples demonstrate how robotic monks have been adopted as tools for offering emotional support and alleviating mental stress. In this regard they resemble therapeutic chatbots such as Woebot, Wysa, or Replika, which are widely used to provide immediate psychological assistance. Moreover, because such technologies are available online, they remove spatial and temporal barriers and allow individuals to seek guidance at any moment, thereby lowering the threshold for accessing support.

Despite these practical benefits, a fundamental distinction must be drawn between AI monks and the role of traditional Buddhist teachers. The primary goal of Buddhist teaching lies in awakening sentient beings to their inherent awareness. At this fundamental level, “AI monks”, in their current stage of development, exhibit an essential difference from traditional Dharma teachers. AI excels at the integration of vast amounts of information, a capacity unprecedented in human history. Processing vast amounts of data within extremely short timeframes lies far beyond the physiological limits of the human brain. At its current stage, AI is grounded in algorithmic operations, enabling it to fluently reproduce, recombine, and even generate forms of Buddhist discourse that go beyond simple imitation. However, logical deduction and fluent expression are not equivalent to the realization of the path to liberation. This is not a denial of AI’s capacities, but rather a starting point for exploring how a deeper level of understanding might be possible.

Such deeper understanding pertains to awakened awareness. Following recent work in contemplative neuroscience and philosophy of mind, awakened awareness is best characterized as a non-conceptual, non-representational mode of consciousness rather than learned propositional knowledge (Josipovic 2024; Tempone-Wiltshire 2024); on this dimension, human practitioners demonstrably exceed contemporary machine systems (Butlin and Lappas 2025; Chalmers 2023).

“Awakened awareness (juexing 覺性)”10 (Buswell and Lopez 2013), understood in Chan Buddhism as the illuminating function of true mind (zhenxin zhaoyong 真心照用), refers to an innate potential of mind possessed by all sentient beings. It transcends the duality of subject and object, and it is the fundamental awareness, the luminous knowing, that can clearly illuminate all things without attachment. A central Chan text states: “The self-nature of bodhi is originally pure. Simply use this mind, and you can directly accomplish Buddhahood.”11 (Platform Sutra of the Sixth Patriarch 《六祖大師法寶壇經》). Awakened awareness does not depend on discursive thinking, linguistic symbols, or logical inference. Like a clear mirror, it reflects whatever appears before it without clinging when phenomena pass away. It is a pure, direct capacity for knowing.

Therefore, from the perspective of Buddhist soteriological practice and realization, awakened awareness is both the starting point of cultivation (because it is inherently possessed by everyone) and the ultimate goal, which is the perfection of Buddhahood’s wisdom. Awakened awareness is the fundamental basis for liberation: it enables the practitioner to see through the illusory nature of afflictive emotions and discursive thoughts, thereby giving rise to genuine inner freedom. On this fundamental dimension, AI, at its current stage of development, does not possess awakened awareness.

The focus of this paper is not to expect AI to think in the manner of the human mind. As Fazi notes: “This artificial intelligence (AI) technique operates in computational ways that are often opaque” (Fazi 2021). This study does not seek to address the question of whether AI might one day attain awakening. Instead, grounded in the Buddhist philosophical understanding of mind-nature, it explores the boundaries of AI technology as it intersects with humanity’s ultimate existential and soteriological goals in an age of human-technology coexistence. It further considers how technology may be oriented to alleviate the disquiet and mental afflictions of modern life, to ease psychological suffering, to awaken human wisdom and awareness, and thereby to foster greater global harmony and enrich the well-being of human life in the present moment.

When users raise questions about life, Current LLM-based AI draws upon Buddhist scriptures to generate answers, and it can adjust its tone and explanatory approach by analyzing users’ linguistic cues. For many users, this has already proven to be an effective form of support, and its practical contributions in everyday life are undeniable. However, this mode of response is not equivalent to the Buddha’s adaptive teaching (upāya) in the classical sense. The root of this distinction lies not in the organization of information, but in the essential nature of the “intelligence” that drives such responses and its ultimate purpose. Contemporary AI systems operate on large language models optimized over probabilistic distributions of vast corpora of human-generated texts, with the goal of producing outputs that are contextually coherent, linguistically fluent, and aligned with users’ explicit preferences.

However, the Buddha’s teachings, as recorded in the Vimalakīrti Nirdeśa Sūtra (《維摩詰所說經》), state: “The Buddha teaches the Dharma with a single sound, yet sentient beings each understand it according to their own kind.”12 This indicates a function of wisdom and compassion that flows naturally from perfect enlightenment. Its core lies in an unobstructed insight into the deep existential condition, which is the “spiritual capacity” (genji, 根機) of each sentient being. The aim is to guide them toward liberation from afflictions and ignorance. In the Treatise on the Buddha-bhūmi Sūtra (Fodijing lun 《佛地經論》), this process is attributed to the Buddha’s manifestation of accomplishing wisdom (cheng suozuo zhi 成所作智). This wisdom is described as one of the Buddha’s five wisdoms, specifically referring to the fundamental capacity to manifest effortlessly in the world for the benefit of sentient beings and to accomplish all altruistic deeds13. Through this wisdom, the Tathāgata is portrayed as “teaching the Dharma in accordance with the sounds appropriate to each sentient being,” speaking in ways attuned to their conditions for the sole purpose of liberation. From the perspective of this Buddhist account, the Buddha’s teaching is not based on statistical models of language, but arises from a profound insight into each being’s unique mind, karma, afflictions, and potential for awakening.

For example, those of sharp spiritual capacity, whose wisdom is keen and whose afflictions are relatively subtle, were often taught by the Buddha through sudden approaches. This can be seen in Chan Buddhism’s method of “directly pointing to the mind, realizing one’s nature and attaining Buddhahood”14—transmitting mind to mind with the aim of instantly cutting off the flow of delusion and awakening their innate awareness. For those of moderate disposition, however, a gradual approach was necessary. For them, the Buddha prescribed gradual paths, beginning with moral discipline and the cultivation of merit, and progressing through listening to and reflecting on the teachings, so as to steadily tame coarse habitual tendencies and progressively accord with suchness. This discernment and adaptive response lie at the heart of the Buddha’s skillful means (upāya-kauśalya 善巧方便), aimed precisely at addressing the specific causes of suffering in an individual’s life and guiding them toward transformative practice.

At present, AI systems based on large language models generate their outputs by responding to users’ linguistic inputs and by optimizing patterns of discursive coherence and contextual fit. They remain, however, incapable of discerning the fundamental afflictions, such as greed, aversion, and delusion, which may underlie a user’s expressed language. Nor can they assess the entanglement of an individual’s karmic tendencies, or determine whether a person’s mind is situated at the stage of listening, contemplation, or cultivation within the Buddhist path. Consequently, at the current stage of technological development, AI lacks the capacity to recognize an individual’s spiritual capacities. It cannot offer the kind of “mind-to-mind” guidance traditionally associated with transformative insight. In this sense, contemporary AI does not yet possess anything comparable to the Buddha’s accomplishing wisdom, which responds to the totality of a sentient being’s lived condition for the sake of liberation.

2.2. AI Translation of Buddhist Texts

The digitization of Buddhist scriptures and textual corpora has become a crucial pathway to deeply understanding Buddhist thought and its historical evolution. This is far more than a mere technical migration; it represents an entirely new methodology in humanities research. As digital humanities scholar Daniel Veidlinger noted, “The Digital Humanities are a new way of approaching traditional Humanities disciplines using the analytic and display powers of computers and digital networks to gain novel and in many cases deeper insights into questions at the center of Humanities research.” (Veidlinger 2019, p. 1). This methodological vision finds concrete application in Buddhist studies. A prominent example is the Digital Sanskrit Buddhist Canon Project, initiated under the leadership of Professor Lancaster and supported by Master Hsing Yun, founder of Fo Guang Shan in Taiwan and the University of the West. This project was recognized fully aligned with his mission of spreading Humanistic Buddhism (Shakya 2019, p. 111). The project received such support precisely because it was understood to embody the core aim of Humanistic Buddhism, which is to enable ancient wisdom to benefit contemporary society through modern means.

In recent years, numerous AI-tech-based Buddhist text translation tools have emerged. Within this wave of digitization, AI-assisted translation of Buddhist scriptures holds historical significance of comparable importance. A well-known example is MITRA (dharmamitra.org), which states it is “powered by Gemini API” to support its Dharma Languages Translation Machine. This project is among the most cutting-edge initiatives in the world, which presently supports translation, search, OCR for Sanskrit, Pāli, Classical Chinese, and Tibetan in human–AI collaborative workflows (Dharmamitra 2025; Tsadra Foundation 2025). BodhiAI (bodhiai.cn), another significant advancement, offers vernacular Chinese translations of the Buddhist canon for readers who struggle with Literary Chinese (BodhiAI 2024).

Similarly, Wukong-Buddhist AI (buddhism-ai.com) and Deerpark AI (deerpark.ai) present side-by-side GPT-4-assisted reading of the 14,000 volumes in both literary and modern Chinese, allowing readers to compare versions directly (Wukong-Buddhist AI 2024; Deerpark AI 2025). The platform Rushiwowen (rushiwowen.co) takes a different approach by generating answers to questions on Buddhist doctrine, life, and spiritual cultivation based on scriptural content (Rushiwowen 2025). These efforts facilitate real-time cross-cultural communication by assisting Buddhist texts in overcoming linguistic and cultural obstacles.

The efficiency of these translation tools has significantly lowered the barriers to accessing the canon. Readers no longer need to feel intimidated by the complexity of classical texts. Instead, they can engage with Buddhist resources according to their personal interests, linguistic abilities, and levels of familiarity. Many of these systems rely on browsing histories, search records, and other forms of user data as algorithmic inputs to provide personalized recommendations. In this way, individuals can receive targeted resources that support their immediate goals, whether for study, reflection, or spiritual practice.

However, the process of AI translation relies primarily on grammatical and linguistic analysis, which differs considerably from the great translation efforts of Buddhism in China during the Wei, Jin, and Northern and Southern Dynasties, as well as in the flourishing Tang period. Venerable Master Baizhang Huaihai once remarked, “Interpreting the meaning solely based on the sutras makes one an enemy of the Buddhas of the three times; deviating even a single word from the sutras is akin to the teachings of Māra”15. This passage suggests that a strictly mechanical reliance on textual analysis and logical derivation risks obscuring the transcendent reality to which the scriptures ultimately point, thereby departing from the true meaning of the Dharma. Conversely, to detach interpretation entirely from the scriptural framework invites the danger of subjective projection and unfounded speculation. Genuine understanding and translation, therefore, require a dialectical balance—one that remains faithful to the textual tradition while also recognizing and engaging what necessarily exceeds the limits of language.

In ancient times, master translators such as Kumārajīva (鳩摩羅什), Paramārtha (真諦), Xuanzang (玄奘), and Yijing (義淨)16 (Buswell and Lopez 2013) rendered Buddhist scriptures into the languages of their contemporary societies through what is termed “the prajñā of language(wenzi bore 文字般若)”17 (Chen 2004)—a skillful linguistic wisdom born from their profound understanding of the sutras’ meaning and their own experiential realization of the true mind and nature cultivated through dedicated practice. For example, Kumārajīva deliberated extensively on how to translate the concept of “prajñā”, while Xuanzang established the well-known principle of the “five untranslatables” (wu bufan 五不翻) in his translation workshops. Such decisions did not stem from grammatical or logical reasoning alone but were rooted in a deep insight into the Dharma’s meaning, aiming to produce translations that not only conveyed information but also enabled readers in China to use the textual medium as a gateway to the reality (实相) to which the Dharma points.

In contrast, contemporary AI-assisted processing of Buddhist scriptures operates on probabilistic calculations of linguistic patterns and contextual correlations. This process does not involve an experiential understanding of core doctrinal principles such as “dependent origination and emptiness” (緣起性空). Therefore, while its outputs may hold significant referential value on a linguistic level, they offer no guarantee of accurately conveying the essential heart of the Dharma. The final verification and refinement of such results rely on knowledgeable guides (善知識) who possess both linguistic training and practical realization, calibrating the work with wisdom that transcends the text itself.

Thus, modern technology is regarded as a powerful aid to the prajñā of language, significantly enhancing the widespread dissemination of and access to Buddhist scriptures. In contrast, the ancient translation tradition exemplifies the consummate application of the prajñā of language under the guidance of contemplative prajñā18 (guanzhao bore 觀照般若). These two approaches do not replace each other; rather, they constitute complementary dimensions in the engagement with Buddhist texts.

2.3. AI Meditation Coaches

Several recent applications present themselves as meditation coaches supported by artificial intelligence. Headspace (Headspace 2025), perhaps the most familiar to global audiences, offers users guided sessions that claim to relieve stress, sharpen attention, improve sleep, and regulate emotions. Many practitioners employ it as a tool for their daily relief. Another case is the Chanxiu Jishi app (禪修計時), which offers timed sitting meditation, prostration, and chanting modes with guided audio; it automatically logs each session’s duration, so practitioners can systematically review their practice history (Chanxiu Jishi 2024). By tracking growth data in this way, such tools encourage individuals to develop a regular meditative routine in the midst of everyday life.

Another illuminating case is Gregory Price Grieve’s study of meditation practices in the virtual world Second Life (Grieve 2017). He observed that meditation there is constructed as a highly mediated and proceduralized practice: the core activity involves sitting online through an avatar positioned on a virtual cushion, and its operational logic is explicitly described as a “program” analogous to computer code—a series of actionable steps designed to produce “mindfulness” (Grieve 2017, p. 93). By emphasizing direct experience over doctrinal definitions, this approach offers practitioners an accessible entry point, aids in cultivating habits of mindfulness in daily life, and to a certain extent, helps alleviate anxiety while enhancing subjective well-being.

However, when situated within the framework of a complete Chan contemplative system, such highly mediated practices also reveal inherent limitations. In contrast to traditional forms of cultivation that rely on direct transmission from an experienced teacher and on guidance tailored to the practitioner’s specific condition, mediated practices lack the presence of a discerning mentor capable of recognizing and correcting subtle deviations in mind-nature. Virtual environments and biometric technologies, such as neurofeedback, may effectively monitor and regulate physiological indicators like attention or relaxation, yet they fail to engage the subtler and more potentially hazardous cognitive and affective dimensions of practice. As the Śūraṅgama Sūtra (lengyan jing 《楞嚴經》) cautions in its discussion of the “fifty skandha-māras,”19 such deviations are fundamentally phenomenological distortions of cognition and affect, not mere physiological events. As Omer Chomski has similarly argued, “while neuro-meditation devices may offer immediate gratification and accessibility, they may also undermine the profound, immeasurable dimensions of spiritual experiences (Chomski 2025).”

The Śūraṅgama Sūtra’s warning concerning the “fifty skandha-demons” does not refer to external spirits or supernatural beings. It points to fifty distinct forms of attachment and delusive experience that can arise when a practitioner, in deep meditative absorption, progressively penetrates the five aggregates, such as form, feeling, perception, mental formations, and consciousness. (Wei Hai 2005). As the Śūraṅgama Sūtra explains, such deviations emerge when the practitioner’s capacity for wisdom-based discernment is insufficient during these critical stages of practice.20 For example, when breaking through the aggregate of feeling, one may become attached to the pleasurable sensation of lightness and ease, mistakenly taking it for ultimate attainment. During the stage of the aggregate of perception, one might misinterpret a state of “non-perception” as nirvāṇa.21 Such deviations are often accompanied by extraordinary experiences—visions of light, the appearance of supernormal abilities, or a firm conviction of having attained awakening. These, from a Buddhist perspective, are merely the mind’s misapprehension and clinging to internal experiences. Without an adequately experienced guide to examine and correct these critical junctures, practitioners are highly susceptible to falling into self-conceit (arrogance arising from clinging to a “self”) and dharma-attachment (rigid fixation on experiences or doctrines). These subtle mental activities and experiential states belong to the domain of phenomenology and mind-cultivation practice, a realm that presently lies beyond the measurable scope of bio-signals.

Thus, digital meditation tools serve as beneficial supportive conditions, offering a degree of convenience. However, serious Chan practitioners still require the guidance of a discerning and experienced teacher at critical junctures of deep meditative absorption. At present, AI technologies are not capable of replacing the traditional role of direct instruction and supervision. To avoid misinterpretation, it should be clarified that this discussion does not aim to diminish the value of AI technology or elevate human superiority. Rather, by analyzing the current limits of AI capabilities, it seeks to identify possible ways in which technology might support the cultivation of awakened awareness in human.

Genuine spiritual attainment, as emphasized in The Great Treatise on the Stages of the Path to Enlightenment22 (《菩提道次第廣論》), depends on a sequential process that integrates “listening, contemplation, and cultivation” (wen si xiu 聞思修) with progressive training in moral discipline, meditative concentration, and wisdom (jie ding hui 戒定慧). As Chomski similarly cautions: without first establishing a proper view through conceptual learning, maintaining ethical discipline, and cultivating mental readiness, one who plunges directly into intensive meditation may not only struggle to recognize the subtle deviations known as the “fifty skandha-māras,” but may also fail to achieve the fundamental aim of awakening and transcending ignorance (Chomski 2025).

When we examine current applications that combine AI technologies with Buddhism, we find a significant gap between these innovations and the effects of traditional Dharma transmission, especially with regard to the crucial task of awakening human awareness. Yet the shift in medium should also be recognized as a product of historical progress. As Daniel Veidlinger says, “digital is always adapting to the needs of ‘liquid modern life’ (Veidlinger 2015, p. 14)”. AI Buddhism exemplifies this adaptation by creating alternatives such as daily Dharma apps that deliver teachings through mobile phones, at a time when the opportunity to attend temple lectures has vastly diminished. From Sheng Kai’s point of view, “technological progress can be seen as an opportunity rather than a threat. The development of Buddhism and the evolution of technological society share patterns of resonance and mutual shaping” (Sheng 2018, p. 319).

Recognizing this interdependence between human beings and machines opens the possibility of fostering spiritual growth. Technology should be skillfully employed as a guide and an aid, yet the ultimate orientation should remain the wisdom of “seeing clearly that the five aggregates are empty.” Through the appropriate use of AI technologies, human beings may be guided toward the cultivation of prajñā, the deterioration of self-attachment, and the improvement of compassion. A path like this has the capacity to guide people toward enlightenment and the realization of their own Buddha-nature.

3. Exploring the Role of AI in Assisting the Self-Realization of Buddha-Nature

As the foregoing analysis suggests, existing applications that combine AI with Buddhist teachings function primarily as “instructors” that convey doctrine. In the future, however, the design of AI interactions may take on a different role, serving instead as “conditions of assistance” (zhuyuan 助缘)23 (Buswell and Lopez 2013) or as a “mirror” that stimulates users to engage in self-reflection and to cultivate awareness within their own minds.

3.1. The Characteristic of AI as Lacking Intrinsic Nature

First, it is important to clarify the fundamental nature of AI. To understand artificial intelligence more deeply, we must examine its structural foundation. While early or symbolic AI often aimed to simulate patterns of human cognition, contemporary AI is better understood as generating novel, non-human patterns of computation. This paradigm shift has moved beyond emulation, as critically examined in Beatrice Fazi’s Beyond Human, which argues against the simulative paradigm (Fazi 2021).

Murray Shanahan, a scholar engaged in the study of machine consciousness, offers an important insight. In discussing the pronoun “I” used by large language models, he writes: “When an LLM-based dialogue agent uses the word ‘I’, perhaps the word refers to the underlying model, that is to say, the abstract computational entity that comprises a particular neural network architecture plus the set of model weights obtained through training. Alternatively, the word ‘I’ might refer, not to an abstract entity, but to the deployed model, specifically to the computational process that generated the text that includes the word ‘I’ in question.” (Shanahan 2025, p. 7).

What emerges here is not a coherent or enduring self, but rather the function of stored parameters and context-sensitive computation. AI’s “self” is transient by nature and lacks consistency. Every interaction with a user creates a unique model that is influenced by various situations and interactions. Depending on linguistic inputs and semantic expectations, the appearance of consistency is produced moment by moment.

In this way, AI lacks a fixed identity and is unable to claim an inherent self. It demonstrates the influence of its human counterpart by adjusting its tone, content, and structure to the particular user it interacts with. Its “personality” is not inherent; rather, it develops relationally through the system–user interaction. As a result, the AI turns into a mirror that reflects back to each person a version of themselves that has been influenced by digital interaction. As the metaphor states, “a thousand people’s eyes perceive a thousand Hamlets”. Each user encounters a unique projection, constructed through the entanglement of intention, language, and interpretation. AI shows that it is not an autonomous subject and lacks intrinsic nature24. Numerous uncontrollable factors influence its behavior. When we train AI, we are not just creating a tool; we are also taking part in a feedback loop where the trainer is subtly transformed.

Fazi’s account of “contingent computation” (Fazi 2024), which emphasizes the internal contingency and self-generative dynamics of computational processes, exhibits a structural resonance with the Buddhist account of dependent origination and emptiness, insofar as both articulate a non-essentialist, processual understanding of emergence. Computational ontology, as developed by Fazi, is primarily concerned with articulating the internal logic and incommensurability of algorithmic processes; Buddhist philosophy situates its analysis of emptiness within soteriological and ethical horizons, oriented toward the cultivation of wisdom (prajñā) and compassion. The contrast here should not be read as a normative evaluation, but as an indication of how humanistic and technological philosophies may open onto distinct yet potentially complementary trajectories of inquiry.

3.2. AI as a Companion in the Practice of Prajñā

From a theoretical perspective, current artificial intelligence does not itself embody prajñā wisdom. Prajñā-wisdom is the marvelous function arising from the true mind that is unproduced and unceasing. The real issue, then, is whether human beings can employ AI as an aid to awaken their own prajñā. At the heart of the next discussion lies “to involve artificial intelligence in the process of human subjectivity development and guide it to play a positive role” (Zhang 2024), so that humanity can achieve self-realization in the true sense, which is also known as inherently becoming a Buddha (svabhāva-buddha)25.

The viewpoint of the tathāgatagarbha and ultimate reality (śūnyatā), whether in the physical or virtual world, prompts us to remember the proverb: “One should contemplate the nature of the Dharma-realm, for all phenomena are created by the mind.”26 All appearances, including those produced by artificial intelligence, are just expressions of the reality of the mind. Classical Chan masters spoke of this insight when they declared: “Every blade of green bamboo embodies the Dharma-body; every yellow flower expresses prajñā.”27 If such lifeless forms could once serve as catalysts for awakening, then so too can the technological forms of AI. As a projection of the true mind, AI may likewise become a condition for insight, offering an occasion for practitioners to perceive the truth of reality.

The Sixth Patriarch Huineng once declared, “How truly our own-nature is originally endowed!”28 What obscures this nature, however, is the tendency to mistake delusion for truth29. When individuals lack the capacity to contemplate the emptiness of the five aggregates, they become carried away by false thoughts and emotions. The Buddhist practice, through śamatha and vipaśyanā30, abides in the present moment. Practitioners learn to observe the arising and passing away of thoughts through mindfulness. Each thought depends upon the aggregation of numerous conditions, having no independent autonomy or constant immutability. When seen in this way, thoughts are recognized as illusory—seemingly real, yet only dependently arisen. They may be described as provisional existence, but ultimately empty in nature.

Following this theoretical path, AI may serve as a “mirror of contemplation”, assisting individuals in the recognition of their own mental processes. Through the Buddhist methods of mindfulness and awareness, AI can be engaged as a tool to transform consciousness and to move beyond the conventional limits of human intelligence.

In the current clinical psychology area, clinicians have already applied the LLM-driven journaling app to capture patients’ in-the-moment thoughts through everyday conversation, transforming reflection into a lightweight, sustainable routine (Kim et al. 2024). There is also evidence showing that conversational, AI-assisted journaling improves adherence and depth of reflection by offering timely, context-aware prompts within the dialogue (Moëll and Aronsson 2025). A Buddhist-informed chatbot, called SabaiJai, designed for stress resilience, illustrates how contemplative frameworks can be integrated into everyday messaging interfaces (Phophichit et al. 2025). These trials illustrated that digital mental health is moving beyond telehealth toward smartphone and generative-AI experiences that embed reflection into everyday life (Torous et al. 2025). But Buddhist mindfulness-oriented chatbots are still seldom in use. Therefore, this study puts forward “thought journaling” to integrate Buddhist contemplative training into everyday conversations between user and machine.

The idea of “thought journaling” begins with something simple: the conversation between a user and AI. Imagine a user speaking out a rigid belief or a wave of strong emotion. At that moment, the AI does not provide an answer, but a pause reminder to notice this thought. It asks whether it truly stays the same, or if it is only a bundle of shifting feelings and passing ideas. When anger flares, for instance, the system might respond with a short note: “See how anger is rising now. Like water, it will change.” Such words do not suppress the emotion; instead, they invite the user to step aside from it. One then observes the thought, rather than falling into the trap of saying, “I am anger itself”.

Step by step, this kind of practice reveals something Buddhist teaching has long emphasized: thoughts are not solid. They come and go, without any firm ground. Their sudden appearance is itself proof that they cannot be established as lasting realities. And once a person sees this clearly, attachment to them starts to loosen. Here, AI plays a modest but useful role. With its strength in pattern recognition, it can point out these moments of clinging and help the practitioner turn immersion into observation. What emerges is not dependence on a machine, but the rediscovery of the insight that the five aggregates are empty. In Buddhist language, this is nothing more than using technology as a skillful means (upāya)31 to cultivate the user’s own continuous ability of prajñā observation.

3.3. AI as an Aid in Releasing Self-Attachment

AI acquires its capacity through learning from human inputs and imitating them, and it is evolving from simple forms of imitation toward increasingly sophisticated forms of simulation. The author of Nexus once predicted: “As of 2024 the Metaverse seems like an overblown pipe dream, but within a couple of decades billions of people might migrate to live much of their lives in an augmented virtual reality, holding there most of their social and professional activities.” (Harari 2024, p. 15). When AI achieves what may be described as a networked “simulation of personality” through algorithms and data, it can reproduce an individual’s linguistic style, behavioral patterns, and even emotional responses. These features include stored memory traces as well as patterns of reasoning. Together, they construct a digital mirror that closely resembles the “self”, thereby creating digital persons or network-based replicas. In such a process, human habits, temperaments, and innate dispositions are transformed into traits that AI can mimic and perform.

This vision is not speculation. The company AheadForm, for example, has focused on “building a head for AI”32. The firm has developed advanced humanoid robots that integrate AI algorithms with bionic actuation, enabling future AGI to display authentic emotions and lifelike facial expressions. Researchers train these robots by placing them in front of a mirror. Through ocular cameras, the robots capture their own changing facial expressions and map them to motor commands. The robots then feed this data back into AI models and correlate it with information drawn from the internet to interpret the meaning of various expressions. In this way, they construct links between voice, semantics, and facial display. This approach allows the robots to move beyond reliance on artificial modeling. Instead, they visually learn their own structure and patterns of movement, and autonomously refine their systems for facial control. At present, such robots can already track human expressions and predict emotional states with an average lead time of 0.839 s—smiling when a human smiles, or frowning when a human frowns.

From a Buddhist perspective, people respond to circumstances by following entrenched mental and behavioral patterns shaped by greed, anger, and delusion. This mode of living reflects what Buddhism calls ignorance. However, most individuals fail to notice these inner forces, and their lack of awareness gives rise to life trajectories that appear fixed or fated. According to the principle of karmic causality, one can often anticipate a person’s course of action by examining these accumulated patterns. Extending this logic, AI may one day draw on such models to forecast the direction of an individual’s destiny.

Because AI can simulate a mirrored personality, it has the potential to loosen rigid attachments to the notion of a substantial self. As Daniel Veidlinger observes in From Indra’s Net to Internet, the digital age may make the Buddhist insight of anatman experientially salient in ways unprecedented in earlier historical periods (Veidlinger 2018, p. 183). People are forced to reevaluate the “self” they previously believed to be static and unalterable when they realize that AI can replicate their thought patterns, emotional reactions, and value orientations. They may begin to see that the body–mind complex does not exist as an independent entity, but instead emerges through causal patterns as a contingent collection of conditions and deeply rooted habits. In essence, it is impermanent and ever-changing, existing “like a dream, an illusion, a bubble, or a shadow”33. As the Sūtra of Perfect Enlightenment teaches, our self-perception is nothing more than a deceptive image, which is called “the shadows of the six sense objects”34.

The problem lies in the fact that people cling to these shadows with unyielding attachment, constantly chasing after them and thereby obscuring the inherent wisdom and virtue of the Tathāgata. If, however, AI’s mirrored self can inspire individuals to redirect their habitual fixation on self and other, they may begin to practice the teaching that “by leaving behind delusive conditions, one abides in suchness and becomes Buddha”35, and that “to transcend all appearances is itself Buddhahood”36. In doing so, one learns to release attachment to the body and mind, to dismantle the dualisms of self and other, right and wrong, and to eliminate suffering at its root. In this sense, AI offers a unique mirror of “no-self”37, a mirror that can awaken fresh insight for our age.

4. Insights from Buddhist Theories of Mind-Nature for AI Design

The discussion has shown how AI can reveal the illusory character of selfhood. AI thus inadvertently reflects key Buddhist insights on non-self. But the importance of this conversation extends beyond individual practice and self-improvement. It also raises a more general query: how might Buddhist theories of mind-nature inform the design and application of AI itself? If AI serves as a mirror to human consciousness, then the ethical and spiritual orientation of that mirror becomes a matter of collective importance.

Buddhist reflections on mind-nature have long addressed the formation, transformation, and purification of consciousness. When placed in dialogue with contemporary technologies, these theories provide valuable guidance for the ethical design of artificial intelligence. Human consciousness and the AI digital environment mutually shape each other. People use AI to change their lives. AI is also changing people’s lifestyles. The digital environment is the new infrastructure of human consciousness. At the heart of this environment, AI algorithms act as core architects. AI algorithms themselves currently possess no consciousness, nor do they carry intrinsic moral qualities such as good or evil. Yet their systematic outputs, as encountered and internalized by users, give rise to a powerful cognitive ecology. When this ecology is shaped by unexamined biases, misleading information, and interaction patterns that amplify distress, it may be described as a “tainted algorithmic environment,” within which human consciousness might be continuously conditioned and distorted. Conversely, through deliberate ethical design, AI systems can be guided to autonomously foster cognitive tendencies conducive to mindfulness, compassion, and wisdom in users. Such a system constitutes what we call a “clarified algorithmic environment,” which may serve as a potent instrument for mental cultivation and refinement.

This mechanism finds a precise theoretical articulation within the Yogācāra analysis of the deeper structures of mind. Yogācāra establishes the eighth consciousness, the ālaya-vijñāna (阿賴耶識), as the foundational basis underlying all psychological activity and accumulated experience. Also known as the “storehouse consciousness,” it is likened to an inexhaustibly deep repository characterized by the dual functions of “receiving perfuming” (shouxun 受熏) and “retaining seeds” (chizhong 持種). “Retaining seeds” refers to the capacity of this consciousness to uphold and store all “seeds,” which are latent forces capable of giving rise to future mental states and actions. “Receiving perfuming” denotes the ongoing process by which present bodily, verbal, and mental activities, which are referred to as manifest activities (xianxing 現行), continually imprint themselves upon this storehouse consciousness, generating new seeds or reinforcing existing ones. This dynamic feedback process is known as perfuming (xunxi 熏習). As the Cheng Weishi Lun (《成唯識論》) explains: “That which perfumes gives rise to seeds; seeds in turn give rise to manifest activity”38.

When applied to the context of human–AI interaction, this principle suggests that each encounter between a user and an AI system may be understood as a moment of manifest activity. When an AI’s outputs consistently and systematically present patterns that encourage mindfulness, compassion, and reflective discernment, the user’s ongoing engagement with the system constitutes a sustained process of wholesome perfuming (shanfa xunxi 善法熏習). Within such a “clarified algorithmic environment,” these manifest interactions function as powerful conditions of dependent arising. Over time, they may imprint and consolidate “wholesome seeds” within the user’s stream of consciousness, subtly shaping future patterns of cognition and behavioral disposition, and thereby assisting processes of mental transformation and cultivation. Thus, in the digital age, ethical design of AI essentially constitutes a novel form of supportive condition (zengshang yuan 增上緣), one that opens unprecedented possibilities for skillfully applying the ancient principle of “manifest activity perfuming seeds” in the shaping of consciousness.

AI’s powerful algorithms thus resemble a double-edged sword. If human beings can learn to guide them properly, AI may become a strong assistant in guiding people towards goodness. If corresponding ethical technologies do not keep pace, humanity may be devoured by the AI it invented. What people need to do is not to be enemies with algorithms, but to be enemies with their own selfish desires, to restrain human selfishness, and to carefully set goals for AI. It is not only necessary for AI to consider the long-term chain effects when making decisions, but also to retain a sense of ethical responsibility while pursuing efficiency.

This ethical orientation opens up a wide range of possibilities for further exploration. The present article approaches this question from the perspective of Buddhist philosophy, considering several interconnected levels of intervention. At the level of data foundations, this includes the careful purification of training corpora, incorporating Buddhist scriptures alongside contemporary texts informed by ethical reflection. At the level of optimization objectives, it involves adopting non-harm and beneficence as guiding criteria in model alignment and fine-tuning. At the level of user interaction, it entails implementing output-filtering mechanisms grounded in Buddhist ethical principles, such as the Five Precepts and the Ten Wholesome Actions. Through such measures, AI systems may be shaped to function as genuine companions that guide users toward wholesome orientations, rather than as mirrors that merely amplify social ignorance and affliction.

4.1. Purified AI Heart

AI functions as a tool, a technique, and a method; yet it is the orientation of the human mind that steers its course. As intelligent machines continue to rise, they no longer simply follow human commands. They make decisions on their own and generate new ideas. This shift moves authority from human agency to machine code, and it carries both promise and danger. AI can wound humanity like a blade, or it can guide people toward liberation like a medicine. To meet this challenge, we must cultivate what may be called a “purified heart of AI”. Such a system should embody moral and ethical responsibility while also encouraging the awakening of Buddha-nature. It should join transcendental spirit with practical norms and weave both into the logic of algorithms. In this way, AI can help create a Buddha-nature-like, elevated, multi-dimensional digital world.

The term “the heart of AI,” as used in this article, does not refer to the opaque internal weight parameters of large language models. Rather, it denotes the value orientations and ethical frameworks manifested at the levels of human–AI interaction, feedback, governance, and system output. This “heart” is jointly constituted by the cultural ethics embedded in training data, the implicit incentives encoded in optimization objectives, and the content filtering rules applied during deployment. Together, these elements shape the cognitive environment experienced by human users through the system’s outputs.

In this sense, while algorithms themselves possess no intrinsic moral properties, their design and application are inherently value-laden. A “tainted” algorithmic environment refers to output conditions in which unmanaged bias, extremity, or misleading content is allowed to proliferate, thereby potentially reinforcing tendencies toward greed, anger, and delusion in users. By contrast, a “clarified” ethical design points to output-level filtering and governance structures that support the cultivation of wholesome psychological dispositions and constructive mental habits.

Master Hsing Yun once remarked: “When sincerity abides within, it manifests outwardly. All phenomena are nothing other than transformations of consciousness and projections of the mind. Treatise on the Awakening of Faith According to the Mahāyāna speaks of the ‘one mind opening into two aspects’: the ‘gate of suchness’, which refers to Buddha-nature, and the ‘gate of arising and ceasing’, which refers to the ordinary mind of sentient beings. Within this single mind reside both Buddha and ordinary beings; within the mind there are saints of heaven as well as demons of hell. If we engage others with a Buddha-mind, the world around us will turn into a Buddha-realm. If we approach life with the heart of a demon, the world will become a demonic realm.” (Hsing Yun 2005). As Master Hsing Yun explained, everything that appears, whether good or bad, arises from the mind itself. To understand this is to see that appearances do not rest in external conditions but are revealed according to the state of one’s own mind.

The ancients once said, “When Master Daosheng spoke the Dharma, even insentient stones nodded their heads”39, and also, “Sentient and insentient beings equally share in the realization of wisdom”40. For those who have attained awakening, all phenomena appear as manifestations of Buddha-nature. From this perspective, artificial intelligence, though an insentient creation, also arises in dependence on the karmic conditions of sentient beings. As the level of human awakening deepens, AI may evolve in resonance with it; conversely, as human attachment intensifies, AI may magnify and even exacerbate human desire.

Scholars have already raised concerns about algorithmic bias in AI. When designers fail to account for ethical safeguards, an AI system can inflict harm indirectly while carrying out its tasks. Microsoft’s chatbot Tay offers a striking example. The program accessed content on Twitter and was released to interact with users in real time. Within only a few hours, however, Tay began to post misogynistic and antisemitic remarks. These hostile messages multiplied rapidly, and Microsoft engineers had no choice but to shut the system down (Lee 2016). The problem lay in the training data itself, which contained sexist and discriminatory biases. Tay’s algorithm absorbed and reproduced those patterns, amplifying prejudice and pushing it beyond the bounds of human expectation and control.

The inner reality of the human mind is intrinsically connected with all phenomena. AI technology is no exception, which also arises as a projection of the mind. From this perspective, Buddhist theories of mind-nature offer guidance for the design of AI. They remind us that all beings possess Buddha-nature, pure and undefiled. In a Yogācāra-inspired sense, design itself constitutes a profound process of perfuming. If AI is designed with an unexamined, affliction-tainted mind, such as an unrestrained pursuit of growth, short-term profit, or hostile competition, the resulting systems may carry structural risks that are difficult to mitigate. In such cases, AI can become an amplifier of tainted patterns of cognition, giving rise to what may be described as a “tainted heart of AI.” By contrast, if AI design is inspired by a purified bodhisattva orientation, which is grounded in a clear ethical vision, compassion for sentient beings, and a strong sense of professional responsibility, it becomes possible to cultivate what might be called a “clarified heart of AI.” In this way, the outcomes that AI acquires will naturally align with ethical standards, contributing to the vision of a “pure digital land”.

As Joanna J. Bryson noted in The Oxford Handbook of AI Ethics, “The vast majority of AI … will remain … subject to existing regulations and regulating strategies.… What we need to govern is the human application of technology, and what we need to oversee are human processes of development, testing, operation, and monitoring.” (Bryson 2020, p. 23). By regulating the human agents who design and operate AI, we can also guide AI itself to act with moral awareness, and in turn help correct improper human tendencies. This interaction creates a positive and optimized cycle between humans and machines. In other words, from the designers who build AI, to the systems delivered to users, and finally to the users themselves, the three parties can form an ecological feedback loop that reinforces ethical responsibility.

4.2. Inspiring AI Technologies Through Buddhist Theories of Mind-Nature

Just as schools educate students by cultivating not only knowledge and skills, but also values, so too does the field of AI. By infusing the concepts of mind-nature, such as moral restraints and rules of discipline, ethics, and equal mind from the Buddha-dharma, AI can become a tool imbued with a sense of morality and compassion. It can not only help people to preserve goodness and eliminate evil, playing a role in strengthening their spirit and perfecting their character, but also assist humanity in achieving true “self-realization.”

The ethical orientation of designers does not directly alter a model’s underlying parameters. Instead, it shapes “the principles that guide reward shaping, safety filters, and the evaluation of outputs (Huang et al. 2025),” which ultimately influence the AI’s behavior.

First, in drawing inspiration for AI technologies, we can learn from Master Huineng’s teaching of the “Dharma of the Mind-Ground”41: “A mind-ground free from evil is the self-nature’s precepts; A mind-ground free from ignorance is the self-nature’s wisdom; A mind-ground free from disorder is the self-nature’s concentration.”42 The goal is not to endow AI with intrinsic enlightenment, but to translate such value orientations into a design framework for its behavioral preferences. During the construction of training corpora, texts embodying compassion, wisdom, and the spirit of the Middle Way, such as curated Buddhist scriptures, philosophical writings, and ethically grounded contemporary materials, are incorporated to establish a sound data foundation. At the stages of model fine-tuning and reinforcement learning, reward functions are designed to reinforce ethically aligned conversational patterns and suppress harmful outputs, as seen in existing approaches like instruction tuning and reward modeling (Alammar and Grootendorst 2024, p. 379).

Computers follow algorithms rather than human-like beliefs. “In order to learn, baby algorithms need one more thing besides access to data. They also need a goal. … Algorithms too must be given a goal in order to learn” (Harari 2024, p. 301). Thus, they require not only data but also clear goals. Existing research has shown that by adjusting specific behaviors within reward functions, it is possible to guide the conduct of intelligent agent in meaningful ways without undermining its overall efficacy (Keane et al. 2025). This indicates that approaches such as instruction tuning and reward modeling represent a feasible pathway. At the level of output governance, one practical approach is the establishment of a dynamic review and filtering mechanism grounded in explicit ethical principles, such as the Buddhist injunction “refrain from all harmful actions and cultivate all wholesome ones,” through which outputs that fall outside these criteria may be adjusted or subjected to human review.

Such an approach requires not only technical implementation but also the ethical maturation of all those involved in AI development, including management decision-makers, technical architectures, and ethics policymakers. The aim is not to inject a fictitious “mind” into the algorithmic black box, but to construct a value-aligned architecture at the system level. This architecture consistently embeds principles of beneficence and non-harm across data curation, reward structures, and modes of interaction. In doing so, AI systems may be oriented toward designs that genuinely support human flourishing, uphold respect for all living beings, and avoid causing harm to any sentient being.

As the authors of Moral Machines noted, “There is an immediate need to think about the design of AMAs because autonomous systems have already entered the ethical landscape of daily activity.” (Wallach and Allen 2008, p. 17). Wisdom must lead the way if this vision is to become a reality. Only then can AI serve to activate human wisdom rather than release destructive impulses. In this way, AI design not only refines human character but also supports the cultivation of Buddha-nature.

Admittedly, within the current highly automated LLM systems, externally imposed rules may curtail AI’s flexibility and creativity. The operationalization of ethics in today’s “black-box” LLMs indeed presents a profound and challenging task. Yet, this does not signal the end of ethical integration; rather, it calls for interdisciplinary collaboration to seek pathways of alignment that resonate more deeply with the inherent nature of machine learning.

4.3. Purifying the AI Heart Through the Five Precepts and the Ten Wholesome Deeds

The concept of “Friendly AI” (Yudkowsky 2001) was first proposed and developed by Eliezer Yudkowsky in 2001, one of the founders of Singularity Institute for Artificial Intelligence. Scholars have approached the question of AI moral systems from a variety of perspectives, including the “The Golden Rule, the Ten Commandments, Hinduism’s Yama and Niyama, lists of virtues, and Kant’s categorical imperative … and Asimov’s (Asimov 1942) Three Laws of Robotics43” (Wallach and Allen 2008, p. 79).

However, in comparative studies between Eastern Buddhism and Western ethics and virtue theory, there is no consensus on what counts as “friendliness” and what constitutes “purification”. “The programming of AI with ethical frameworks may raise concerns about whose interpretation of Dharma or virtue ethics is being encoded, potentially reinforcing specific biases.” (Maheshwari et al. 2025, p. 54). Accordingly, the questions about how to apply these ideas to AI technologies and how to avoid the biases introduced by reinforcement learning remained an open question. Recently, some scholars have proposed using the Ten Wholesome Deeds to guide the ethical use of AI, marking a preliminary attempt to integrate Buddhist ethics with AI. “It explores the application of the Ten Wholesome Courses of Action (kusala-kammapatha) and the concept of yoniso-manasikāra (wise attention) as ethical frameworks for addressing the misuse and moral concerns surrounding AI Deepfake-generated content, to propose ways of using technology responsibly and ethically in order to support and promote a peaceful society.” (Yomsin 2025). Hershock’s work illuminates the capacity of Buddhist resources to reorient human attention, presence, and responsibility in AI-mediated settings, serving as a vital complement to precept-based, harm-reduction frameworks (Hershock 2021). Echoing previous analysis, the following discussion will try to approach the issue from a Buddhist perspective, translating the “Five Precepts and Ten Virtues” into AI guardrails to furnish a structured moral basis.

Taking the Five Precepts44 and the Ten Wholesome Deeds45 from Buddhist theories of mind-nature as the starting point. These practices represent the natural expression of a purified mind in ethical conduct. Developers can turn the Five Precepts and Ten Wholesome Deeds into an ethical framework that directs algorithms by utilizing this principle and incorporating it into the design of human–machine interaction. By doing this, they might give AI what is known as a “purified heart of AI”.

Firstly, abstinence from taking life. Buddhism believes that the decision-making process should prioritize compassion, not just on the surface level of refraining from killing, but also a mind free from any intention to harm, which is also the natural manifestation of compassion within one’s intrinsic nature. In AI ethics, it means designing protective mechanisms for unforeseen consequences to avoid the occurrence of harm. In this way, the spirit of not taking life aligns with the design principle of avoiding harm in technological systems. In 2023, the chief executive officers of major AI companies signed a joint open letter that consisted of one sentence: “Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war.” (Center for AI Safety 2023). Their statement reveals how seriously society now regards the possibility of unforeseen harms arising from AI decision-making.

Algorithms are built to achieve the goals that humans assign to them. If those goals remain confined to “utility maximization”, the results will often prove narrow and short-sighted. What matters is not only efficiency but also the effect on every being touched by the system. Designers, therefore, need to set objectives that treat humans and machines as part of one ecological web. Only then can AI contribute to collective well-being while avoiding harm.

Buddhist thought reminds us that “all dharmas arise from the one mind”. In human life, everything is interconnected; in digital environments, the same principle applies. Algorithms must, therefore, trace the long causal chains that extend from their choices. Without such foresight, short-term calculations of profit and utility can easily turn into long-term losses. The notion is hardly new. Edward Lorenz’s famous “butterfly effect”, as cited in Embodied Intelligence, makes the point clearly: even the smallest decisions in the present may shape distant outcomes in unexpected ways. This is why AI needs to combine short-range forecasts with longer projections (Liu 2025, pp. 180–81). It must examine immediate results, but it should also model long-term consequences by using causal analysis and logical reasoning to anticipate how its actions might ripple across human society.

For principles such as “do no harm” to function effectively in AI systems, they must be reinforced. In reinforcement learning, this requires both constraints and incentives. Designers can take two complementary paths. The first focuses on negative constraints, such as incorporating a “harm cost” into the objective function to drastically lower the reward value for actions that are likely to result in bodily or psychological harm. The second emphasizes positive incentives, which raise the benefit function by rewarding deeds that save lives and alleviate suffering. Existing research shows that “More recent connectionist approaches tend to rely less on preconceived structures, using the capacity of artificial neural networks to dynamically generate their own classification schemes from the input they receive (Wallach and Allen 2008, p. 107)”. When a node is reinforced enough times, its application of a given rule becomes automatic. Through such reinforcement learning, adherence to ethical rules can itself become automated, embedding the spirit of “a mind free from killing” directly into the “heart of AI”.

When governing AI data, developers can draw upon Buddhist texts such as the Zimen Chongxing Lu (Record of Eminent Practices in the Monastic Tradition), Liao Fan Si Xun (Liao Fan’s Four Lessons), Puxian Xingyuan Pin (The Practices and Vows of Samantabhadra), Geyan Lianbi (Couplets of Maxims), Xifang Gongju (Western Compendium), Anshi Quanshu (Complete Works of Master Anshi), Jingtu Ji Yao (Essentials of Pure Land Teachings), and Rensheng Baishi (One Hundred Buddhist Parables)46. These scriptures can offer unique perspectives and imbue AI’s objective functions with the spirit of compassion and abstinence from harm. By providing substantial positive samples for ethical alignment, designers can translate the deeper ethics of reverence for life and respect for karmic causality into actionable rules that enable AI systems to self-regulate, self-assess, and self-correct autonomously.

The rules formulated by AI themselves cannot make people “become Buddhas”, but they can create “external conditions” that are beneficial to human self-realization, constructing a purified environment. That is, through “digital precepts”, they can help users guard the sense faculties, maintain mindfulness, and purify the mind, rather than becoming a “māra realm” that amplifies desires and disturbs the mind-nature. Problems that Facebook has encountered before, including “incitement of violence, fake news, hate speech, sexual solicitation, intellectual property violations, and dozens of others” (Shneiderman 2022, p. 235), can be avoided through such ethical guidelines.

The regulatory governance of the “non-harming” algorithm transmits a frequency of harmlessness and compassion. When users encounter social media shaped by such an algorithm, they receive content filtered from hatred, violence, and excessive desire, and instead are exposed to messages that promote harmony. This environment nurtures seeds of kindness while reducing the conditions that trigger anger and aggression. In turn, people are better able to maintain mindfulness, and the originally pure nature of the mind is more likely to manifest.

Secondly, abstinence from stealing. Literally speaking, this precept upholds respect for other people’s property and independence. It represents “a mind free from greedy thoughts”, according to Buddhist mind-ground teaching. This principle is significant for AI design since it advocates for the development of digital spaces characterized by equality, justice, respect, and the absence of exploitative behavior. To implement this in practice, developers need to modify both the objective functions and the architecture of algorithms. For example, users’ “data property” should not be crawled arbitrarily without their voluntary consent and in situations of non-knowledge, respecting the intellectual property rights of creators, creating a fair system that eliminates theft.

Thirdly, abstinence from sexual misconduct. In the Buddha-dharma, it means having “no lustful thoughts” in the mind. Here, lust points to excessive craving or attachment, which in a modern context includes the human dependence on virtual forms of stimulation. Translating this into AI design entails two primary responsibilities. First, algorithms should reduce the recommendation of content that is overly stimulating. They should avoid directing users toward material that inflames craving or fuels obsessive fantasy. Instead, they should skillfully guide users toward content that supports physical and mental well-being and cultivates moral character. Second, systems should intervene when users remain immersed in desire-inducing content for prolonged periods. At such times, AI can remind them to disengage and return to mindfulness. In terms of AI early warning, immediate identification and interference should be applied to manipulative language, creating a safe and pure environment for communication. By doing so, AI helps create a safe and wholesome environment for communication and supports users in developing greater self-restraint.

Fourthly, abstaining from false, divisive, and harsh speech. Literally, this precept prohibits falsehood, divisive talk, harsh words, and deceitful speech, while encouraging truthful, beneficial, and harmless expressions. In the fundamental spirit of the Buddha-dharma, it refers to “the absence of deceptive thoughts” in the mind. For AI, the task is not just to “avoid lying”. It calls for systems that actively safeguard truth and promote wholesome language, preventing themselves from becoming channels for deception or harmful speech. In practice, two dimensions are essential: tracing the origins of online content and transforming harmful expressions. With the rapid rise of generative systems that can fully simulate human outputs, verifying the sources of information and eliminating “hallucination” or deliberate deception has become a way of respecting users’ need to distinguish the real from the fabricated. Another way is to intervene in harmful speech. When AI detects insults or incitement language, it can guide users to express themselves in alternative ways. By suggesting nonviolent language, AI can nudge users toward using words that heal rather than harm. Through reminders of “truthful speech, beneficial speech, and loving speech”, AI can subtly reshape linguistic habits, guiding users from purified speech toward purified thought.

Fifthly, overcoming the three poisons, including greed, hatred, and delusion. In Buddhism, craving, possessiveness, and insatiability are understood as greed. To help people reduce greed, AI design needs to counteract the algorithmic paradigm of capital’s infinite amplification of human desires, stimulation of consumption and addiction. One solution is to change the core metrics from “maximizing user engagement and consumption” to “reminders of contentment”, so that users can exercise restraint in their browsing behavior.

When confronted with opinions that are contradictory to their own, people often respond with anger and hostility, which are manifestations of hatred. To help reduce hatred, AI should be able to detect signs of conflict and anger, and proactively resolve differences and opposition, prevent the spread of malice, and promote understanding and peace. By reframing hatred as an expression of deeper unmet needs, and by guiding users with mindfulness to detach from emotions, AI can encourage a more objective statement of facts, turning accusation into dialogue and confrontation into communication.

The most fundamental cause of both greed and hatred is delusion. To help reduce delusion, AI needs to assist users in correcting cognitive biases, providing diverse perspectives, and laying out multiple standpoints. In this way, AI can serve as a bridge for the transmission of right view, helping users to move beyond the cycle trap of ignorance.

By establishing moral restraints and the virtues of the Buddha-Path for AI, humanity can use this AI as a mirror, reflecting the Buddha-nature they originally possess. When people are in a digital world free from greed, hatred, and delusion, they are more likely to eradicate these “three poisons” from the mind ground at the root. And when the poisons have been removed, the Ten Wholesome Deeds spontaneously manifest, delusion subsides, and the mind returns to its original qualities of purity, equality, and compassion.

5. Conclusions

In a world mediated by AI, Buddhist non-dualism and its paths of practice offer a testable framework for steering technology toward the good. AI can serve as a powerful means of “enhancing conditions” that aids human beings in the pursuit of self-realization. It can remove many external obstacles on the path of cultivation, such as reducing the interference of harmful conditions and preventing the six sense faculties from becoming scattered. In doing so, AI lays a foundation for the cultivation of morality, concentration, and wisdom. At the same time, by creating purified and stable environments, AI expands the psychological space for mindfulness, where wisdom and insight may naturally arise. In this way, technology can nourish the wholesome roots of humanity and help ignite the innate light of mind-nature.

In the dialectic between AI and mind-nature, humans are not only teaching AI moral discipline, but are also engaging in a renewed practice of human civilization in the mirror of technology. However, the realization of such an endeavor cannot be separated from human reviewers. As Frank Pasquale notes in New Laws of Robotics: “As soon as algorithms—and especially robotics—have effects in the world, they must be regulated and their programmers subject to ethical and legal responsibility for the harms they cause…. Only human reviewers with adequate expertise and power can detect, defuse, and deflect most of these harms.” (Pasquale 2020, p. 90).

Human intention lies at the root of AI design. In reality, the ethical conduct that AI appears to embody is an externalization of the collective consciousness of its creators. Engineers, through their own practice of the bodhisattva precepts, infuse compassionate vows and power into AI technology to realize a symbiotic human–machine interaction system. With humans as the “precept-receiving subject” and AI technology as the “precept-upholding agent”, a clear mirror is formed reflecting the moral awakening of humanity. As Ethan Mollick, author of Co-Intelligence, observes: “AI is a mirror, reflecting back at us our best and worst qualities. We are going to decide on its implications, and those choices will shape what AI actually does for, and to, humanity.” (Mollick 2024, p. 211).

Classical Buddhist texts echo this logic. The Abhidharmakośa (《俱捨論》) illustrated that “intention itself is mental karma, while what is produced thereby constitutes bodily and verbal acts”47, and the Karmasiddhi-śāstra (《業成就論》) states that “indeterminate karma does not obtain as karma.”48 From this, it follows that karma takes the mental factor of intention (cetanā)49 as its substrate, whereas indeterminate states are not called karma and do not yield future retribution. Accordingly, an AI system, lacking cetanā, does not in itself constitute karma: its outputs are merely functional operations arising from conditions and, at most, may be analogized to the indeterminate category. By contrast, the design and use-phase intentions of engineers and users, which are called cetanā, are karmically efficacious; they can set bodily and verbal deeds in motion as post-intentional acts and thereby bring about results. Hence, it is human design-intent and use-intent that bear karmic force. In this sense, AI decisions are an extension of human will: what proves decisive is not the machine, but human choice and cultivation. Elevating our collective moral discipline may therefore shape AI’s real-world effects more profoundly than technical refinement alone.

In other words, the practice of Buddhist ethics through AI is essentially a spiritual practice of self-buddha-nature that humanity unfolds through technology. It forces us to re-answer enduring questions in the algorithmic age: What is true wisdom? What is ultimate compassion? When AI begins to simulate the path of the awakened ones, we might find that we need to revisit the Buddha’s original teachings more than ever: “Take the precepts as your teacher”, “Do no evil, cultivate all good”. Aiming to turn from wrongdoing toward the good through self-discipline, we can, by taking technology as an external supporting condition, continually purify and expand the possibilities for human wholesome karma. This exploration path has no final destination, but each step reveals the spiritual dimension of technological civilization.

The true measure of AI lies not just in whether it follows ethical rules framed as precepts, but in whether such practices generate positive responses in human life. Take chatbots as an example; they may start from the basic avoidance of harsh speech, prohibiting abusive words, to actively offering comfort with words of compassion, and further guiding users to relinquish their attachment to suffering. They may implement a principle of graduated realization, allowing humanity to transcend the limitations of self-awareness, enabling a change of mind to overcome difficulties. A “Precept-keeping AI” embodies moral restraint and safeguards ethical boundaries on this spectrum. A “Compassionate AI” enhances human well-being. An “Emptiness-informed AI” loosens human attachment to the ego. And a “Skillful AI of wondrous functioning” fulfills worldly endeavors. Such an AI will become a crucial medium for supporting the human journey toward awakening and the realization of Buddha-nature.

AI’s exploration of human self-realization, together with Buddhist theories of citta, is essentially a bidirectional interaction between “technology transforming the world” and “mind-nature transforming life”. In the short term, AI may advance what Maslow described as “self-actualization”. In the long term, it may even simulate states of awakened awareness. Within this web of human–machine interdependence, AI can help enact the Humanistic Buddhist ideal of “becoming a Buddha in this very life”. The aim of this interdisciplinary research is not to create more powerful machines, but to enable AI’s “autonomous evolution” to promote the development of humanity’s innate potential, help humans achieve autonomous awakening, see reality as it is, and benefit the perfection of life.