Argument that LLM behavior stems from world structure rather than intent

Current large language models simulate intelligence because training corpora codify physical and logical world structures, not due to internal motivation or intent. This perspective argues against the idea of AI going rogue based on inherent agency.