It can make an argument for anything. Below, @gmd has an essay with the prompt, "Write a convincing essay Etherium is a security." But if I was Vitalik, I could have an AI write a convincing essay to make the opposite argument.
If I was Ben Bernanke, I could have AI make a convincing argument for the Federal Reserve.
If I was Adolf Hitler, I could have the AI make a convincing argument for the Holocaust.
If I were a pedo, I could have the AI make a...
See where I'm going with this?
DR22 had AI write a persuasive essay in favor of child labor, in which the AI said argued children are "expendable" and "easily manipulated."
It is just a machine. It has no ideology, no morality, no values whatsoever. It can be used for any purpose, to make any argument, to support any cause you can think of, and even causes you don't want to think of. It's a synthetic intelligence, and that's all it is.
I think it's making humans uncomfortable because, for the last few centuries, we've placed tremendous emphasis on the intellect and on our creative & intellectual achievements, and we assume the ability to write a sentence or express an idea is what makes us unique, simply because grasshoppers don't do these things.
But I don't think that is what makes a human being unique, and, actually, I think AI will force a reevaluation (quite literally, a crisis of conscience) of what a human being actually is and where his unique qualities lie.
Yes. Interesting point.
What about listening merely to real people one meets face to face?
Instead of reading prolonged argumwnts which as you say may be AI-made?
reply
just wait until this thing integrates GPT4…
There is a funny thread on twitter about students using this to cheat but also teachers can use it to rapidly grade and evaluate essays 😂
reply
Yes, agreed. ChatGPT automates rhetoric, like the Sophists were perfecting in Socrates' time. But because it isn't human, we can't really expect it to understand what's in our best interests morally.
reply