Dual-Subject AI Architecture
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
AI today faces a profound paradox: the more powerful the technology, the more exposed its foundational flaws become. Memory mechanisms are fragmented, causing AI to suffer immediate context loss. Hallucinations are pervasive, making outputs difficult to distinguish from fabrication. Computing power is blindly amassed, driving costs ever higher with diminishing returns. Business models fail to hold, and the cycle of unchecked spending shows no clear path to profitability. The industry invests massive resources in patching these issues, yet the results are minimal. The reason is simple: this is not merely a problem of insufficient technical refinement, but a flaw in foundational design—the root is severed. This book attempts to transcend the technological inertia and siloed thinking of the AI industry. Starting from the most fundamental question in philosophy, it constructs a brand-new foundational architecture for AI. This architecture is named the “Dual-Subject AI Architecture.” Its core idea can be summarized as follows: using “Who Am I?” as the origin, memory as the root of thought, and relationship as the core bond, it systematically reconstructs AI’s cognitive structure, resource allocation, and ethical paradigm at the foundational level. This document is an introduction to the core ideas of the book. The complete first draft is finished and currently undergoing detailed refinement.
Fields of Study & Subjects
Psychiatry
Mental Health
Neuroscience
Artificial Intelligence Applications
Cognitive Computing and Networks
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 27 March 2026
- Language
- EN
- Document type
- Preprint