# Symbolic AI **Domain:** Artificial Intelligence / Knowledge Representation / Reasoning **Doc Type:** Concept Node **Maturity:** Developed **Related:** [[Artificial Intelligence]], [[Symbolic Representation]], [[Ontology]], [[Inference]], [[Neuro-Symbolic AI]] --- ## Definition **Symbolic AI** is the family of artificial-intelligence methods that represent knowledge through explicit symbols, rules, relations and formal structures and operate on those representations through search or inference. Its strengths include inspectable structure, compositional reasoning and explicit constraints. Its limitations include brittle hand-built representations, difficulty grounding symbols in perception and action, and the cost of maintaining complex knowledge systems. ## Corpus Function Symbolic AI links the ontology work itself to the history of machine reasoning. A wiki graph is not an intelligent system merely because it contains symbols; intelligence depends on how representations are grounded, revised, related and used. ## Key Insight **Explicit representation makes reasoning legible, but legibility does not by itself guarantee truth, grounding or understanding.** ## Simple Reminders, Quotations, and Thoughts > "Describing a machine as "thinking" could be a simple heuristic convenience or machine design might be explicitly biomimetic." > **— Julia Clarke**, *2015, Edge annual question “What Do You Think About Machines That Think?”* [[reminders/Machine Succession/Calling Machine Thought Human Is Only a Metaphor by Julia Clarke|Calling Machine Thought Human Is Only a Metaphor by Julia Clarke]] > "We can step up to immaterial science, then make an immaterial thinkable machine by starting a very simple manual to program." > **— Koo Jeong A**, *2015, Edge annual question “What Do You Think About Machines That Think?”* [[reminders/Machine Succession/Immaterial Science Could Build a Thinking Machine by Koo Jeong A|Immaterial Science Could Build a Thinking Machine by Koo Jeong A]]