3 ms·
You’ve shifted your goal posts. You asked me for a problem I can solve with thinking. Not a problem another human can solve for me by their thinking. I have giv
by randomImmigrant 1mo ago
You’ve shifted your goal posts. You asked me for a problem I can solve with thinking. Not a problem another human can solve for me by their thinking. I have given you exactly what you asked for. An LLM cannot solve it. A human can’t solve it for me but can solve it for themselves (or not be capable of it), it varies.
Whatever you thought you were going to prove with your question, I do hope you’ll dwell on what my example tells you about what thinking is. It isn’t a string of words only. It can be that, but it’s more.
> No reasonable definition of thinking requires that thinking can't be paused or turned off. Please state your point more clearly.
It does. Please define thinking however you want. I’ll show you that it’s a requirement.
> Every problem that previously required human thought can be solved by LLM. QED.
Again, this is the same superficial thinking. Firstly, as my example above showed, no. There are other examples. It requires human thought to cook. No LLM can, nor is smell digitizable in an equivalent way to humans, and yet, cooking is one place where having a human nose is determinative to making appetizing food. Your QED is logically absurd, but is also resting on an assertion that’s flatly untrue.
> No. How many times do I have to explain that the mechanism of thinking is not the same, just as the mechanism of producing light is not the same? You need to think to understand this, which is something that frontier LLMs have no problem with, but which most humans struggle with.
When your “proof” for thinking is the output, I will also compare it to the output, light. My analogy shows the idiocy of your argument, and you don’t want to sit with its implications.
LLMs replicate human language. And many tasks human language can do, LLMs can do or help with. Thinking is not just producing word strings. We think about our family. We think about our bodily aches and pains. We think about how best to do various physical tasks. All this we do sans words, and LLMs don’t do. To call their internal dynamics “thinking” is silly wishful thinking by those who don’t have the ability to go read and understand biology.
And it’s a sad statement of how little you think of your own mind that you won’t explore facts that are outside your expertise that will actually teach you how your mind works, and the wondrous complexity underlying it, and the potential it has to keep doing new things.
I’m going to stop engaging at this point. You seem convinced LLMs think. I wish you the joy of that belief, whatever it is. Religion is always absurd, and this new one is no different.
- lern_too_spel 1mo ago> You’ve shifted your goal posts. Remember the problem we're discussing is whether these LLM systems can think. Your deliberate misinterpretation of a question does not make you win that argument. An LLM can solve problems for itself in just the same way that a human can solve problems for itself. It cannot directly rewire your thinking in the same way that another human cannot. > It requires human thought to cook. Why? What part of cooking requires human thought? https://www.moley.com/ https://www.moley.com/ > nor is smell digitizable in an equivalent way to humans, Sensors aren't thinking, but people are working on better olfactory sensors too. https://www.nature.com/articles/s44182-025-00071-y https://www.nature.com/articles/s44182-025-00071-y > assertion that’s flatly untrue. You have yet to show me a problem that can only be solved by human thinking. > When your “proof” for thinking is the output What other way can we show thinking? From the input? Obviously not. > My analogy shows the idiocy of your argument It shows the idiocy of your argument. The output is thinking, just as the output is light. You are the one who continues to mistake the output for the mechanism. > Religion is always absurd, and this new one is no different. Your assertion of magic in biology that can't be replicated artificially is indeed absurd and one that almost no scientist would agree with.