AI Hallucinations: Why Large Language Models Make Things Up (and How to Fix It)

Re:AI Hallucinations: Why Large Language Models Make Things ... by JKCalhoun on Wednesday December 04, 2024 @10:00PM (#42324372)

I won't defend the term but am curious what you think would have been also concise but more accurate. Calling them for example "inevitable statistical misdirections" doesn't really roll off the tongue.

[ Parent | on HN ]

Re:AI Hallucinations: Why Large Language Models Make Things ... by malfist on Wednesday December 04, 2024 @11:15PM (#42324906)

It's a bug. Any other system where you put in one input and expect a certain output and get something else it'd be called a bug. Making up new terms for AI doesn't help.

[ Parent | on HN ]

Re:AI Hallucinations: Why Large Language Models Make Things ... by thomasfromcdnjs on Thursday December 05, 2024 @12:57AM (#42325374)

Need not I say different bugs have many names...

[ Parent | on HN ]

Re:AI Hallucinations: Why Large Language Models Make Things ... by lolinder on Thursday December 05, 2024 @08:38AM (#42328125)

I actually disagree with bug for the same reason I disagree with hallucination: it creates the idea that there's an error in processing that needs to be fixed, rather than an inherent characteristic of the computing paradigm that needs to be managed.

To be accurate, a term would need to emphasize that it's describing an opinion about the output rather than something that happened inside the program. "Inaccuracies" would be one term that does that fairly well.

[ Parent | on HN ]

Re:AI Hallucinations: Why Large Language Models Make Things ... by goatlover on Thursday December 05, 2024 @02:37AM (#42325877)

Confabulation, if the desire is to use a more apt psychological analogy.

[ Parent | on HN ]

Re:AI Hallucinations: Why Large Language Models Make Things ... by alan-crowe on Thursday December 05, 2024 @04:50PM (#42333206)

Statmist

[ Parent | on HN ]