Re:AI behavior guardrails should be public
by AlecSchueler on Wednesday February 21, 2024 @04:01PM
(#39459501)
A lot of such patterns are actually because of systemicly racist patterns present in the training data though, so it's somewhat unavoidable.
[ Parent
| on HN ]
Re:AI behavior guardrails should be public
by alpaca128 on Wednesday February 21, 2024 @04:05PM
(#39459544)
It's not unavoidable, but it would cost more to produce high quality training data.
Re:AI behavior guardrails should be public
by AlecSchueler on Wednesday February 21, 2024 @04:58PM
(#39460176)
Yes, somewhat unavoidable.
Re:AI behavior guardrails should be public
by _heimdall on Wednesday February 21, 2024 @05:53PM
(#39460751)
Do we have enough info for to say that decisively?
Ideally we would see the training data, though its probably reasonable to assume a random collection of internet content includes racist imagery. My understanding, though, is that the algorithm and the model of data learned is still a black box that people can't parse and understand.
How would we know for sure racist output is due to the racist input, rather than a side effect of some part of the training or querying algorithms?