Here's a good starting point for those of you interested in learning more: https://www.nature.com/articles/d41586-018-05707-8 … (Nature is one of the top science journals in the world)
-
-
Show this thread
-
Everyone in my replies going full-Kavanuagh and crying/yelling in opposition to the reality of how AI training actually works look like Flat Earthers
not a single one actually arguing with facts or reason.Show this thread
End of conversation
New conversation -
-
-
This Tweet is unavailable.
-
This Tweet is unavailable.
- Show replies
-
-
-
imagine thinking that once something uses math it can't be wrong imagine having math and numbers be that much of a black box mystery we gotta improve primary education, folks
-
This Tweet is unavailable.
- Show replies
New conversation -
-
-
It's even more simple than that. Take for example a loan: A POC is likely to have a lower income, less credit, unexceptional occupation/work history. This alone is a formula/algorithm that compounds the divide between those who are privileged & marginalized, white & POC.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
There is no world without biases! It's about picking the preferable biases like the bias for desirable over undesirable outcomes, the bias to preserve nature or the bias that collaboration is better then egotism. Culture is a sack of biases so the time to reflect on them is now!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
The worst part is that the algorithms are opaque and controlled by a few large tech companies. We can change the bias when the data maintains its privacy and can still be analyzed.
@oceanprotocol -
In principle AI bias could be controlled better than human judgement bias, even for opaque AI, but we'd actually have to do that.
- Show replies
New conversation -
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.