3 ms·
Glad you posted. A lot of AI code simply cannot be trusted. In particularly, it's pretty obvious that the teams using it either don't do a lot of QA or don't un
by hodgesrm 5y ago
Glad you posted. A lot of AI code simply cannot be trusted. In particularly, it's pretty obvious that the teams using it either don't do a lot of QA or don't understand how to test their systems in the first place. I find it mind-boggling that this technology is so widely used with such limited understanding.
The interesting question is how the hell do we fix this?
- jpambrun 5y agoOne example doesn't support your claim that QA was not sufficient. You will always get misclassifications, most won't be offensive to anyone. This is not a medical device or a self driving car, this is Facebook trying to squeeze more engagement..
- hodgesrm 5y agoThis type of misclassification is offensive and for good reason. Similar errors have happened regularly for a long time and in many different systems. [1] The comments about enlarging training sets illustrate the blinders that many technical people have about this technology. There does not seem to be a notion of independent quality assurance to detect what should by now be obvious problems. [1] https://www.bostonmagazine.com/news/2018/02/23/artificial-intelligence-race-dark-skin-bias/ https://www.bostonmagazine.com/news/2018/02/23/artificial-in... Edit: oops, left out source
- jrm4 5y agoI have never seen a more sadly oblivious statement than this: the suggestion that Facebook squeezing engagement is significantly less dangerous than malfunctioning medical devices or self driving cars. Facebook is more dangerous; precisely because the harm in getting it wrong is so much easier to hide.