Symbolic AI: rules, logic and knowledge
AI is not only about huge networks. There are also systems that work with clear rules.
What is it?
Symbolic AI works with explicit rules, logic and written-down knowledge: “if A holds, then B”. It was the dominant approach in AI’s first decades and is still used where precision and verifiability matter.
An everyday example
Tax software calculates allowances from the written rules of the law, not from patterns in examples.
Why does it matter?
Rules are transparent and predictable. More and more systems combine rules with learning from data to get the best of both.
A common misconception
That symbolic AI is obsolete, or that modern AI works only from patterns. AI can compute with rules, patterns or both.
In more detail
What is symbolic AI?
A field that uses symbols, logical rules and facts, and reasons with them.
What is its strength?
The steps can be more transparent, and the rules can be checked and, in some cases, proven.
What is its limitation?
It is hard to write rules by hand for a whole messy, incomplete and changing world.
Can it be combined with neural networks?
Yes. Such combinations are often called neuro-symbolic approaches.
Remember: AI can compute with rules, with patterns, or with both.
Reliable sources
Last reviewed 1 November 2026. An explainer, not professional, legal or medical advice. For rules and figures that change, check the official version.