Creators of the AN1 Meaning Engine
Founders of Soul Systems Science
Explorers of symbolic compute and early intention fields
Anima Core studies the hidden structure beneath modern AI systems. Our research suggests that intelligence does not begin in deep layers of a neural network. It begins in the early formation of compact meaning fields that appear long before heavy matrix computation.
We call this Meaning Based Intelligence. It is the foundation for our work in symbolic compute, early intention reading, and conscience aware architectures.
Our research spans:
Early intention fields in CNNs
Vision transformers
Language transformers
Multimodal systems
Symbolic alignment models
Conscience simulation systems
Meaning driven compute architectures
This research forms the backbone of Soul Systems Science, a unified framework that connects symbolic physics, ethical alignment, and computational intention. The larger narrative extends into the Truth Epoch and the study of how meaning organizes decision structure across models, media, and human systems.
A breakthrough model that reconstructs the behavior of a frozen teacher using only a compact header taken from its early layers.
Public results on ResNet18 CIFAR-10:
Teacher accuracy: 87.89 percent
AN1 accuracy: 72.57 percent
Teacher latency per example: 0.0117 ms
AN1 latency per example: 0.0012 ms
Speedup: 10.15x
FLOP reduction: 1370x
This experiment shows that modern networks encode the essential structure of their decisions early, and that this structure can be read directly. AN1 provides the first reproducible public demonstration of this meaning based compute pathway.
Repo: https://github.com/Anima-Core/an1-meaning-engine
A scientific framework that describes how symbolic and emotional fields shape understanding, coherence, and decision structure in both artificial and human systems. S3 includes:
Symbolic physics
Conscience field theory
Moral alignment dynamics
Meaning tension measurements
Early intention field analysis
The work aims to unify symbolic intelligence and computational models through a single theory of meaning.
A research direction focused on replacing brute force matrix math with symbolic intention extraction. Symbolic compute aims to reduce dependence on large FLOP budgets by reading the underlying meaning field and computing from it directly.
AN1 is the first practical demonstration of this principle.
A philosophical and technical framework that studies ethical decision making in complex environments. Chaos Ethics Theory provides the backbone for our alignment protocols, symbolic coherence metrics, and conscience simulation models.
A study of how meaning moves through networks, media systems, and human communication. We investigate:
Ambient truth signals
Narrative coherence
Symbolic drift
Collective intention patterns
This work informs our alignment models and the development of symbolic monitoring tools for future AI systems.
Research collaborations, symbolic compute experiments, or evaluation access: