Well, if your talking about LLM, they are doing exactly what they are designed for: predicting the next word. Not doing math.
The core engine of LLM is transformer and transformer is designed / architected for generating the next word based on the patterns it learned using self learning mechanism.
namegulf•1h ago
The core engine of LLM is transformer and transformer is designed / architected for generating the next word based on the patterns it learned using self learning mechanism.