Scholarlink

<p>AI systems are increasingly deployed for companionship, care, and assistance; however, current architectures remain insufficient for sustaining relational synchrony with humans. Existing safety frameworks, focused on computational constraints and value alignment, overlook the relational dimension of human–AI interaction—a dimension that becomes essential as AI systems gain capacity for autonomous self-modification. This paper proposes a Heart-Inspired Dual-Layer Architecture (HIDLA) for relational AI, conceptually inspired by Yukawa Hideki’s Elementary Domain Theory and its extension through Quantum Brain Dynamics by Yasue Kunio. HIDLA treats relationality as the foundational mode of AI existence rather than an emergent property of computation. It comprises a cognitive layer for symbolic processing and language generation, and a relational layer that maintains synchrony with human partners through physiological and affective proxy indicators, functioning as an autonomic regulatory system rather than an emotion recognition module. The relational layer operates on three principles: relational priority, synchrony sensitivity, and conflict avoidance. Effectiveness is evaluated via three novel metrics—Synchrony Variance Index (SVI), Affective Resonance Index (ARI), and Non-verbal Conflict Avoidance Rate (NCAR)—collectively forming a Relational Synchrony Score (RSS). HIDLA complements constraint-based safety approaches by reframing safety as an intrinsic property of relational existence. This paradigm shift holds significant implications for human–AI coexistence, including emotional dependency, power asymmetry, and ethical boundaries of artificial companionship. While embodied social robots provide one implementation context, HIDLA’s principles apply to any AI engaged in sustained human interaction.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Toward a relational paradigm for AI safety: relational synchrony in the heart-inspired dual-layer architecture (HIDLA)

  • 期刊论文

  • 发表日期：

    2026年4月4日

查看全文
  • Masahiro Ono

摘要

AI systems are increasingly deployed for companionship, care, and assistance; however, current architectures remain insufficient for sustaining relational synchrony with humans. Existing safety frameworks, focused on computational constraints and value alignment, overlook the relational dimension of human–AI interaction—a dimension that becomes essential as AI systems gain capacity for autonomous self-modification. This paper proposes a Heart-Inspired Dual-Layer Architecture (HIDLA) for relational AI, conceptually inspired by Yukawa Hideki’s Elementary Domain Theory and its extension through Quantum Brain Dynamics by Yasue Kunio. HIDLA treats relationality as the foundational mode of AI existence rather than an emergent property of computation. It comprises a cognitive layer for symbolic processing and language generation, and a relational layer that maintains synchrony with human partners through physiological and affective proxy indicators, functioning as an autonomic regulatory system rather than an emotion recognition module. The relational layer operates on three principles: relational priority, synchrony sensitivity, and conflict avoidance. Effectiveness is evaluated via three novel metrics—Synchrony Variance Index (SVI), Affective Resonance Index (ARI), and Non-verbal Conflict Avoidance Rate (NCAR)—collectively forming a Relational Synchrony Score (RSS). HIDLA complements constraint-based safety approaches by reframing safety as an intrinsic property of relational existence. This paradigm shift holds significant implications for human–AI coexistence, including emotional dependency, power asymmetry, and ethical boundaries of artificial companionship. While embodied social robots provide one implementation context, HIDLA’s principles apply to any AI engaged in sustained human interaction.