Uh, no. OpenAI and Anthropic and Google and co really, really, really DNGAF whether or not you understand the code their LLMs write for you. LLMs are not capab

Uh, no. OpenAI and Anthropic and Google and co really, really, really DNGAF whether or not you understand the code their LLMs write for you. LLMs are not capable of reasoning or following code flow. They’re predictors of next tokens. They’re increasingly astonishingly good at predicting next tokens, to the point that they sometimes appear to be reasoning. But they can’t actually reason.