My Vision
Motivation
By understanding all phenomena that naturally emerge under constraints, let us make every being freer.
Only by first fixing the constraints can comprehensive expansion proceed without category errors, and only then can the results safely reach real people.
Part Ⅰ.
A Life Where Everyone Can Truly Understand and Embrace Themselves
Conscious phenomena are expressions that emerge when complex systems achieve self-organization. I explore and seek to understand the metabolic-neural-physical causal processes underlying these expressions — and on that basis, trace how the mutual aggregation of simple reflexive behaviors becomes instantiated as a self-organizing system with genuine causal agency.
This understanding is formalized into a self-awareness and regulation model (designed to address not only affective but also cognitive dimensions) and proposed as a socially institutionalized model for dissemination, in the way SEL has been integrated into educational systems. Through this, each individual strengthens their capacity for self-awareness and self-regulation, with the goal of cultivating sovereign identity robustness, resilience, and problem-solving ability.
Part Ⅱ.
Realizing ASI That Co-evolves With Humanity
To enrich mechanical signals, I explore a new computational representational system — moving beyond binary architecture toward novel devices capable of representing multiple states within a single unit, together with the corresponding ISA redesign. At the hardware level, this aims to expand the range of cases in which machines can semantically understand and express information. At the software level, it seeks to improve computational precision by incorporating relational operators that handle modal logic, extended state spaces, and probability orders beyond those computable within current binary frameworks.
This system, capable of expressing continuous state spaces, holds compatibility with biological signal patterns and serves as the physical foundation for translating between organoid signals and machine representations.
On this basis, the goal is to minimize AI’s most critical failure modes (hallucination, sycophancy, conformity, and reward hacking) and to advance the field of scientific simulation, where such operations are most consequential. Further aims include minimizing misdiagnosis in clinical domains where subjective judgment is dominant, and realizing an ASI that cognitively grows and co-evolves alongside human beings.
Part Ⅲ.
A Time Framework Derived from Irreducible Atomic Physical Constituents, and the Universe from the Perspective of Dissipative Structures and Dynamical Cycles
Dark matter/energy and the modeling of time remain among the most pressing open problems in astronomy. Due to timescale constraints, direct empirical verification is extremely difficult — and precisely because of this, unnecessary variables and domain extensions are often introduced into existing models.
I pursue time modeling by eliminating components that are not re-described as new order parameters or effective degrees of freedom grounded in microscopic constraints. I also explore whether residual terms exist in dark matter/dark energy calculations, and investigate the universe from the perspective in which each celestial body possesses a growth-and-decay lifecycle.