We are nowhere near AGI though. I suspect it is still centuries away. An LLM is essentially autocomplete with some very clever tricks that allow it to do some

We are nowhere near AGI though. I suspect it is still centuries away. An LLM is essentially autocomplete with some very clever tricks that allow it to do some tasks quite well. But it is not even conceptually close to actual intelligence.