I wonder if the issue with the Chekov’s gun example is that it involves a gun. Suspect ChatGPT may have been trained to avoid giving responses in which people get shot or are subjected to other kinds of violence. May be worth trying with a more innocuous prompt.
Well. I wouldn't fault the reasoning; I'd say instead that the second premise is false. As explained later, harmful chemicals are added to cigarettes, and so cigarettes don't satisfy the condition "made of plants" where that is understood to mean "made *only* of plants".
If instead "made of plants" is taken to mean "made of plants plus other things" then it's the first premise which is false.
Useful essay, thanks.
I wonder if the issue with the Chekov’s gun example is that it involves a gun. Suspect ChatGPT may have been trained to avoid giving responses in which people get shot or are subjected to other kinds of violence. May be worth trying with a more innocuous prompt.
"All things made of plants are healthy.
Cigarettes are made of plants.
Therefore, cigarettes are healthy"
Well. I wouldn't fault the reasoning; I'd say instead that the second premise is false. As explained later, harmful chemicals are added to cigarettes, and so cigarettes don't satisfy the condition "made of plants" where that is understood to mean "made *only* of plants".
If instead "made of plants" is taken to mean "made of plants plus other things" then it's the first premise which is false.