"Facebook doesn’t seem to understand that their tools are the bread and butter of manipulators.”https://t.co/Oy900pWbYz pic.twitter.com/vObcSXBgjo
— Rachel Thomas (@math_rachel) July 26, 2018
"Facebook doesn’t seem to understand that their tools are the bread and butter of manipulators.”https://t.co/Oy900pWbYz pic.twitter.com/vObcSXBgjo
— Rachel Thomas (@math_rachel) July 26, 2018
On how a private/closed Facebook support group for sexual assault survivors was used to harass them. (remember that Facebook requires users to use their real names, yet doesn't require groups to disclose who is behind them)https://t.co/Oy900pWbYz pic.twitter.com/jeAxDKDpTD
— Rachel Thomas (@math_rachel) July 26, 2018
Healthcare Machine Learning News (yikes)
— Data Science Renee (@BecomingDataSci) July 26, 2018
h/t @bretcdennishttps://t.co/RRK0VlAsJb
WHOA. IBM Watson has been recommending 'unsafe and incorrect' cancer treatments, according to their own internal documents https://t.co/0Bav6XDRNn pic.twitter.com/HWKLx5KrV3
— Kate Crawford (@katecrawford) July 26, 2018
The third piece in our data ethics series explores the five 'C's as a framework for considering the impact of your product: https://t.co/l97y2e337v
— Hilary Mason (@hmason) July 25, 2018
This was the moment I realized I needed to take the Google AI blog less seriously (even though the posts are written by AI researchers):https://t.co/b5kxp3lmEw pic.twitter.com/JWw3GdWUvw
— Rachel Thomas (@math_rachel) July 25, 2018
The misleading hype around IBM Watson Health was so harmful, and IBM was primarily responsible for that: https://t.co/NB4YrAfHJa
— Rachel Thomas (@math_rachel) July 25, 2018
One cause not mentioned in this article on misleading coverage of AI:
— Rachel Thomas (@math_rachel) July 25, 2018
tech companies themselves (for their misleading press releases, blog posts, and marketing).https://t.co/n7dSButl5e
warnings Facebook execs received in 2013, 2014, & 2015 about hate speech and potential for genocide in Myanmar
— Rachel Thomas (@math_rachel) July 24, 2018
"That’s not 20/20 hindsight. The scale of this problem was significant and it was already apparent."https://t.co/E4R0gHrVDI pic.twitter.com/9bwLqeCB9F
Awesome talk by @math_rachel on recognizing bias in predictive models and AI. H/t @ledellhttps://t.co/MgaRdsTz8L pic.twitter.com/UeLw4gNJ4z
— Karandeep Singh (@kdpsinghlab) July 24, 2018
God dammit — I really don’t want to go through the hassle of changing cloud disk provider, but this is seriously tempting me to shun @Dropbox https://t.co/TvBGwdaxEh
— Thomas Lin Pedersen (@thomasp85) July 23, 2018
Surveillance capitalism is Orwellian when accurate, Kafkaesque when inaccurate.
— Aza Raskin (@aza) July 21, 2018
Health insurance companies use unverified life-style data (e.g. what neighborhoods you've lived in, media consumption habits) to predict health care costs.
/credit to @F_Kaltheuner for the framing https://t.co/PfbUiPkyv2