MINDAS.ME is introducing DCS, a cross-scale causal evolution framework that examines how new levels of organization may emerge across approximately 13.8 billion years, from physical structure and life to mind, civilization and artificial intelligence.

DCS begins with a simple question: how did a universe without life, brains or AI eventually produce systems capable of memory, prediction, planning and deliberate intervention in the world? Rather than replacing the sciences that study each stage, the project asks whether recurring causal transitions can be compared across them.

The framework draws on causal set theory, which describes spacetime as a set of discrete events ordered by causal relations, and extends that intuition into a computable direction: representing causal relations as typed, weighted and vectorized causal arrows, so that complex causal structures can enter graph models, dynamical models and AI computation.

DCS focuses on a proposed sequence in which possibilities are constrained by dynamics, some structures persist, lower-level complexity becomes partly closed or compressed, and new macro-level variables acquire predictive or causal relevance. In simplified form, the program studies a progression from causal structure and persistence to causal compression, causal emergence, causal prediction and causal intervention. Each transition may also involve symmetry breaking: equivalent possible paths branch, part of the lower-level freedom is compressed into stable structure, and new ordered causal degrees of freedom are released.

DCS also distinguishes between pre-biological structural evolution and Darwinian biological evolution. Before life, the relevant processes concern the formation, stability and transformation of physical structures. After heredity, variation and differential reproduction emerge, selection acts on systems capable of generating and transmitting structure across generations.

The project extends this comparison to brains, civilization and AI. Neural systems use information from the past to anticipate future states and guide action. Human societies extend memory, coordination and causal influence through language, writing, institutions and technology. AI systems increasingly combine prediction with memory, tools, software and real-world interfaces, raising a further question: when does greater predictive ability become greater causal reach?

Artificial intelligence is not only a subject of DCS but also part of its research workflow. Large language models and AI agents are used for literature discovery, cross-disciplinary comparison, counterargument generation, fact checking, knowledge organization and iterative review. The work is led by Rongjie Wei through a one-person-company model, which also serves as an experiment in whether AI can give an individual access to parts of the search, synthesis, critique and knowledge-management functions that previously required larger interdisciplinary teams.

"The goal of DCS is not to declare a final theory, but to build an open causal framework that can be challenged, formalized, computed and tested," said Rongjie Wei, founder and CEO of Shenzhen Ruier Maisi Technology Co., Ltd. "AI can expand the scale of questions one person can explore, but scientific value still depends on evidence, mathematics, criticism and falsifiability."

DCS is presented as a developing research framework rather than an established unified theory. Its next stage focuses on operational definitions, mathematical formulation, computational models, links to existing scientific literature, and tests that could distinguish its claims from alternative explanations. The framework was publicly introduced on September 16, 2026 during the global online event "Finding the First Principles of Evolution."

Research materials and the open dataset are available through the project site https://mindas.me/ , the archived record https://zenodo.org/records/22709952 , the Kaggle dataset https://www.kaggle.com/datasets/rongjiewei/dcs-causal-structure-evolution-theory and the model and data space https://huggingface.co/weirongjie .

ABOUT THE COMPANY
Shenzhen Ruier Maisi Technology Co., Ltd. is a Shenzhen-based technology company developing AI-native projects and research initiatives related to human cognition, intelligent systems and emerging forms of human-AI collaboration. MINDAS.ME serves as a public platform for related research, ideas and projects.

FULL POSTAL ADDRESS AND PRESS CONTACT
Shenzhen Ruier Maisi Technology Co., Ltd.
Room 210, Building 5, COFCO Innovation Center, Xingdong Community, Xin'an Street, Bao'an District, Shenzhen, Guangdong Province, China.
Contact: Rongjie Wei, Founder and CEO
Phone: +86 188 2656 2299
Email: [email protected]
Web: https://mindas.me/