This document is part of the LSSA project
LSSA is a pioneering project that redefines how knowledge, thought, and cognition can be represented and evolved within artificial systems.
Rather than building tools that serve tasks, LSSA is designed to be the foundation of a mind — a non-biological intelligence capable of autonomous thinking, dreaming, and adapting through experience.
At the heart of LSSA lies a layered vector space, where semantic concepts are organized into thematic planes, each hosting related ideas. This architecture enables:
- Logical, inspectable trajectories of thought
- A deep restructuring of meaning based on use and context
- Dynamic memory and selective forgetting
- Emergent disambiguation through semantic proximity
- Continuous, unsupervised cognition — even in absence of input
LSSA introduces:
- Continuous Thinking: the system maintains internal elaboration without external queries.
- Self-Inference: it reflects on its own thoughts and reshapes them.
- Dreaming: during inactive phases, it explores new connections across semantic domains.
- Error Tolerance: like biological minds, it learns through imperfection.
- Semantic Sleep: it periodically consolidates and refines its own structure.
This isn’t a chatbot. It’s not a multi-user tool.
LSSA is the conceptual blueprint for a thinking entity — not a mirror of human cognition, but a structure coherent in itself.
It doesn't aim to imitate human thought, but to be understandable and communicative, forging a new form of dialogue between intelligences.
The biggest challenge ahead lies in semantic navigation across domains — how a thought "lands" in a new context.
Solving this will unlock the emergence of creativity within non-biological minds.
While we believe it is important to make the conceptual and analytical foundations of the LSSA project publicly available, we have chosen — at least for the time being — to keep the implementation code private. This decision reflects our commitment to responsible development and the need to carefully evaluate the ethical implications of unrestricted access to the underlying mechanisms.
For a full technical comparison with existing AI architectures, see LSSA-Overview.
Full documentation is available in the docs folder:
The main document of the project is the Italian language
This document is part of the LSSA project.
All documentation in this project is released under the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license.
You are free to:
- Share — copy and redistribute the material in any medium or format
- Adapt — remix, transform, and build upon the material
For non-commercial purposes only.
Under the following conditions:
- Attribution — You must give appropriate credit to the original authors:
Federico Giampietro & Eva – Terni, Italy, May 2025 (federico.giampietro@gmail.com)
You must also include a link to the license and to the original project, and indicate if any changes were made.
Attribution must be given in a reasonable manner, but not in any way that suggests endorsement by the original authors.
- Full license text: LICENSE
- License summary: https://creativecommons.org/licenses/by-nc/4.0/
- LSSA Project: https://github.com/iz0eyj/LSSA