A Spherical Model for General Artificial Intelligence (AGI)
Year: 2025
Type: dissertation
Abstract: Current artificial intelligence (AI) systems exhibit remarkable performance in perception,reasoning, and language. However, despite their functional power, they remain structurallyincomplete. This paper proposes that AGI (Artificial General Intelligence) may still “miss a sphere”— a metaphor for systemic closure and holistic integration. The spherical model is introduced as aconceptual and geometric framework that represents intelligence as a closed, self-consistent, andfeedback-balanced system. By contrasting planar and spherical models of cognition, the paperargues that the key to AGI lies not in more data or parameters but in achieving structural wholeness— the ability to contain self-reference, adaptivity, and boundary coherence within one dynamicsphere. This model offers a new path toward system-level intelligence, potentially bridging the gapbetween mechanical learning and autonomous understanding Author’s Note (added post-publication):The conceptual framework and ideas presented in this work were independently developed by the author.AI tools (ChatGPT) were used solely for assistance in English composition and formatting.
Author: Xie, Xian-Feng
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7103655922
Cites:
Cited by:
Related to: