bias   10565

« earlier    

Amazon's Gender-Biased Algorithm Is Not Alone - Bloomberg
Amazon’s recruiting engine went to great lengths to identify and weed out women. A women’s college in the education section of a resume was an automatic demerit. By contrast, the presence of typically male vocabulary, such as “executed,” was a point in favor. These are just two examples of how computers can sift through data to find proxies for the qualities that they want to seek or avoid. What seems like offhand, irrelevant information correlates to things like gender, race, and class. And like it or not, gender, race, and class are very important in how our world works, so their signals are very strong. This allows computers to discriminate without their creators intentionally doing so.

What makes Amazon unusual is that it actually did its due diligence, discovered the troubling bias and decided against using the algorithm.
algorithm  algorithms  sexism  racism  bias  amazon  job  jobs  research  computers 
yesterday by msszczep
Discrimination in Online Ad Delivery by Latanya Sweeney :: SSRN
A Google search for a person's name, such as “Trevon Jones”, may yield a personalized ad for public records about Trevon that may be neutral, such as “Looking for Trevon Jones? …”, or may be suggestive of an arrest record, such as “Trevon Jones, Arrested?...”. This writing investigates the delivery of these kinds of ads by Google AdSense using a sample of racially associated names and finds statistically significant discrimination in ad delivery based on searches of 2184 racially associated personal names across two websites. First names, previously identified by others as being assigned at birth to more black or white babies, are found predictive of race (88% black, 96% white), and those assigned primarily to black babies, such as DeShawn, Darnell and Jermaine, generated ads suggestive of an arrest in 81 to 86 percent of name searches on one website and 92 to 95 percent on the other, while those assigned at birth primarily to whites, such as Geoffrey, Jill and Emma, generated more neutral copy: the word "arrest" appeared in 23 to 29 percent of name searches on one site and 0 to 60 percent on the other. On the more ad trafficked website, a black-identifying name was 25% more likely to get an ad suggestive of an arrest record. A few names did not follow these patterns: Dustin, a name predominantly given to white babies, generated an ad suggestive of arrest 81 and 100 percent of the time. All ads return results for actual individuals and ads appear regardless of whether the name has an arrest record in the company’s database. Notwithstanding these findings, the company maintains Google received the same ad text for groups of last names (not first names), raising questions as to whether Google's advertising technology exposes racial bias in society and how ad and search technology can develop to assure racial fairness.
data  bias  technology  culture 
yesterday by sew245
Why We Should Expect Algorithms to Be Biased - MIT Technology Review
Technologies driven by algorithms and artificial intelligence are increasingly present in our lives, and we are now regularly bumping up against a thorny question: can these programs be neutral actors? Or will they always reflect some degree of human bias?
data  bias  technology  culture 
yesterday by sew245
Cognitive bias cheat sheet
I’ve spent many years referencing Wikipedia’s list of cognitive biases whenever I have a hunch that a certain type of thinking is an official bias but I can’t recall the name or details. It’s been an…
bias  psychology  reference 
3 days ago by bradbarrish
Why “I’m not racist” is only half the story | Robin DiAngelo - YouTube
"All systems of oppression are highly adaptive, and they can adapt to challenges and incorporate them. They can allow for exceptions. And I think the most powerful adaptation of the system of racism to the challenges of the civil rights movement was to reduce a racist to a very simple formula. A racist is an individual—always an individual, not a system—who consciously does not like people based on race—must be conscious—and who intentionally seeks to be mean to them. Individual, conscious, intent. And if that is MY definition of a racist, then your suggestion that anything I’ve said or done is racist or has a racist impact, I’m going to hear that as: you just said I was a bad person. You just put me over there in that category. And most of my bias anyway is unconscious. So I’m not intending, I’m not aware. So now I’m going to need to defend my moral character, and I will, and we’ve all seen it. It seems to be virtually impossible based on that definition for the average white person to look deeply at their socialization, to look at the inevitability of internalizing racist biases, developing racist patterns, and having investments in the system of racism—which is pretty comfortable for us and serves us really well. I think that definition of a racist, that either/or, what I call the good/bad binary is the root of virtually all white defensiveness on this topic because it makes it virtually impossible to talk to the average white person about the inevitable absorption of a racist world-view that we get by being literally swimming in racist water.

White fragility is meant to capture the defensiveness that so many white people display when our world views, our identities or our racial positions are challenged. And it’s a very familiar dynamic. I think there’s a reason that term resonated for so many people. I mean even if you yourself are to explain white fragility it’s fairly recognizable that in general white people are really defensive when the topic is racism and when they are challenged racially or cross racially.

So the fragility part is meant to capture how easy it is to trigger that defensiveness. For many white people the mere suggestion that being white has meaning will set us off. Another thing that will set us off is generalizing about white people. Right now I’m generalizing about white people, and that questions a very precious ideology, which is: most white people are raised to see ourselves as individuals. We don’t like being generalized about. And yet social life is patterned and observable and predictable in describable ways. And while we are, of course, all unique individuals, we are also members of social groups. And that membership is profound. That membership matters.

We can literally predict whether my mother and I were going to survive my birth and how long I’m going to live based on my race. We need to be willing to grapple with the collective experiences we have as a result of being members of a particular group that has profound meaning for our lives. We live in a society that is deeply separate and unequal by race. I think we all know that. How we would explain why that is might vary, but that it’s separate and unequal is very, very clear.

While we who are white tend to be fragile in that it doesn’t take much to upset us around race, the impact of our response is not fragile at all. It’s a kind of weaponized defensiveness, weaponized hurt feelings. And it functions really, really effectively to repel the challenge. As a white person I move through the world racially comfortable virtually 24/7. It is exceptional for me to be outside of my racial comfort zone, and most of my life I’ve been warned not to go outside my racial comfort zone.

And so on the rare occasion when I am uncomfortable racially it’s a kind of throwing off of my racial equilibrium, and I need to get back into that. And so I will do whatever it takes to repel the challenge and get back into it. And in that way I think white fragility functions as a kind of white racial bullying, to be frank. We make it so miserable for people of color to talk to us about our inevitable and often unaware racist patterns that we cannot help develop from being socialized into a culture in which racism is the bedrock and the foundation. We make it so miserable for them to talk to us about it that most of the time they don’t, right? We just have to understand that most people of color that are working or living in primarily white environments take home way more daily slights and hurts and insults than they bother talking to us about."
racism  oppression  robindiangelo  whitesupremacy  civilrights  race  2018  intent  consciousness  unconscious  morality  whiteness  socialization  society  bias  ideology  fragility  defensiveness  comfort  comfortzone 
4 days ago by robertogreco
How To Kill Your Tech Industry
After the war, British computing breakthroughs continued, and British computers seemed poised to succeed across the board, competing with US technology on a global scale. But by the 1970s, a mere thirty years later, the country’s computing industry was all but dead.

What happened? The traditional history of computing would have you understand this change through the biographies of great men, and the machines they designed. It would gesture towards corporations’ grand global strategies, and the marketing that those companies pushed to try to define what computers were for an entire generation of workers. It would not, however, focus on the workers themselves. And by ignoring them, it would miss the reasons for this catastrophic failure—a failure that remains a cautionary tale for many other countries today, particularly the United States.
history  feminism  business  argument  culture  bias  computing 
4 days ago by kmt

« earlier    

related tags

2018  341webmgmt  351  644  a  accuracy  aggregators  aggression  ai  algorithm  algorithmic-bias  algorithms  allies  allsides  amazon  analysis  analytics  and  anger  argument  article  articles  artificialintelligence  asd  assertive  availability  balance  bbc  behavior  belief  bigdata  bigotry  bogus.metrics  book  books  brain  break  business  can  censorship  ceos  chart  civilrights  clintonhillary  codabr  cognition  cognitive  cognitivescience  comfort  comfortzone  communications  computers  computervision  computing  confederate  confirmation  confront  confrontation  confrontational  consciousness  consequence  conservatism  corbycaroline  cornell  course  court  courtroom  ct  culture  data-analysis  data  death  decision.making  defensiveness  degrees  demographics  department  design  disability  discrimination  distribution  diversity  down  ecj  election  emotion  employment  engineering  ethics  facebook  facialrecognition  fact  fairness  faith  female  feminism  feminist  fernandaviegas  food-stamps  food  foxnews  fragility  fundraising  futurism  gender  google  gop  gov2.0  grants  graphic  group  health  heather  highered  history  how  http  ia  iat  ideas  ideology  ifttt  imaging  immigration  in  inclusion  influence  innovation  instagram  intelligence  intent  internet  irrational  is  joannemcneil  job  jobs  journalism  justice  knowhere  l-m-sacasas  labor  language  law  learning  lexicon  liberal.arts  liberalism  life  list  lists  living  long_paper  looks  macdonald  machine-learning  machine  machine_learning  machinelearning  mask  math  mckinsey  media  medicine  men  mentorship  metrics  mind  misinterpretation  ml  morality  murdochrupert  nabilhassein  netflix  news  newsletter  nick-carr  no  not  numbers  nytimes  obstacles  oppression  over  overcoming  paper  parole  pattern  penalty  people  perception  perceptions  philosophy  photography  photos  pocket  policing  politics  portrait  power  prediction  prison  privacy  probability  problem  problems  propaganda  psychology  pundits  race  racial  racism  recommendationengine  reference  refusal  regression  republicans  research  robindiangelo  science  service  sex  sexism  singal  skills  snap  social-media  socialization  socialmedia  society  sources  special-interests  state  statistics  stats  stem  stereotype  stereotypes  study  subscription  supervisedlearning  supreme  surveillance  surveillancecapitalism  taking  teaching  technique  technologism  technology  telegraphy  tells  test  the  three  threshold  toread  training  transgender  trends2019  trobuleshooting  trust  trustabletech  uk  unconscious  us  usa  usda  video  virginia  voting  webapps  whiteness  whitesupremacy  wikipedia  winning-slowly-season-6  women  work  writing  wsj  yes  概率 

Copy this bookmark: