Symbolic AI, also known as Good Old-Fashioned Artificial Intelligence (GOFAI), is an approach to artificial intelligence that represents knowledge through symbols, logical rules, and structured relationships rather than learning from large datasets.
Instead of identifying patterns statistically, a symbolic AI system reasons through problems using predefined rules, much like following a decision tree. For example, a medical diagnosis system might apply a rule such as “if the patient has a fever, a cough, and difficulty breathing, then they may have pneumonia” to arrive at a conclusion that can be fully traced and explained.
The transparency of symbolic AI is one of its strengths. Because every conclusion comes from an explicit rule, it's easier to understand how the system reached its output and to adjust its logic when needed. This makes it well-suited to domains where knowledge is structured and well-defined, such as expert systems and certain areas of natural language processing.
Symbolic AI contrasts directly with machine learning and deep learning, where models learn from data but offer little visibility into how a specific decision was made, a challenge that explainable AI (XAI) continues to address.