Knowledge in Animals and Machines

Area of Science:

• Philosophy of Mind
• Artificial Intelligence
• Cognitive Science

Background:

• Existing frameworks for knowledge often fail to encompass both biological and artificial systems.
• Understanding knowledge in diverse entities like animals and machines requires a unified conceptual approach.

Purpose of the Study:

• To develop a conceptual framework for knowledge applicable to both animals and machines.
• To engineer a novel concept of knowledge specifically for large language models (LLMs).
• To propose a comprehensive taxonomy of knowledge spanning biological and artificial categories.

Main Methods:

• Drawing upon philosophical theories of knowledge.
• Developing a novel conceptual framework.
• Engineering a new concept of large language model (LLM) knowledge.
• Proposing a taxonomy of knowledge.

Main Results:

• A conceptual framework for knowledge in animals and machines has been established.
• A new concept of large language model (LLM) knowledge has been engineered.
• A taxonomy categorizing knowledge across biological and artificial entities, from C. elegans to LLMs, has been proposed.

Conclusions:

• The developed framework provides a foundation for rigorously accounting for knowledge realization across diverse systems.
• The approach respects the varied biological and physical origins of knowledge.
• This work bridges the understanding of cognition in natural and artificial intelligence.