.@OpenAI controversy, no one (including me) talked about #ethics of using @reddit for training. It is toxic to #women #minorities. I was harassed and threatened on @mxlearn when I spoke up about #NeurIPS @Miles_Brundage @jackclarkSF can u comment on this choice? @samcharringtonhttps://twitter.com/__yoson__/status/1101593216992849920 …
-
-
Replying to @AnimaAnandkumar @OpenAI and
we didn't use reddit data, we used outbound links from reddit ranked at two stars or higher by users. Part of the reason for thinking about different release of this model is precisely because of issues like this (which occur in larger models also). We'll be sharing more here!
3 replies 1 retweet 19 likes -
Replying to @jackclarkSF @OpenAI and
Thanks. I still find it problematic.
@reddit threads have problematic links to all sorts of sites. Right thing to do would have been to choose appropriate data source to begin with. Certainly not have a big press coverage with highly biased model#responsibleAI6 replies 1 retweet 19 likes -
Replying to @AnimaAnandkumar @jackclarkSF and
As
@jackclarkSF mentioned, we hope to share more of our approach to the bias issue very soon. Agree it’s a key concern!1 reply 0 retweets 3 likes -
Replying to @Miles_Brundage @AnimaAnandkumar and
I think that humanity is biased and we should avoid all their data where possible ;-)
1 reply 0 retweets 1 like
No, I am not saying that. Let us all work towards creating better data rather than blindly taking it from internet. @OpenAI can work on such initiatives. Why not scrape text out of feminist media like @BitchMedia I hope someone trains a model from that and see contrast ;)
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.