Reading discussions around Gettier cases, JTB, fallibilism, etc., started feeling like watching the same idea circle itself with slightly fancier synonyms.
At some point you have to ask: What are we actually trying to determine?
If “knowledge” can’t exist without a perfectly objective observer, which does not exist, but every sense and memory we have is fallible, what are we even measuring?
And if you bring that into AI alignment: Are we building “truth detectors,” or are we just trying to formalize human-like confidence and calling it “knowledge” to make it sound smarter?
Reading discussions around Gettier cases, JTB, fallibilism, etc., started feeling like watching the same idea circle itself with slightly fancier synonyms.
At some point you have to ask: What are we actually trying to determine?
If “knowledge” can’t exist without a perfectly objective observer, which does not exist, but every sense and memory we have is fallible, what are we even measuring?
And if you bring that into AI alignment:
Are we building “truth detectors,” or are we just trying to formalize human-like confidence and calling it “knowledge” to make it sound smarter?