Atlas of AI
Embedding identity claims in technical systems as though they are facts observable from the face.
What is digital epidermalization?
The community of people who self-track
Who is the Quantified Self community?
The historical, institutional, social, cultural, people, and material conditions that shape information
What is context?
a procedure for solving a mathematical problem in a finite number of steps that frequently involves repetition of an operation
What is an algorithm?
Working to reduce oppression in data science and algorithms is ___________.
What is justice
The technical AI research community relies on cheap, ______ labor for many tasks that can't be done by machines.
What is crowdsourced/crowdsourcing?
The average or the norm
What is "normal"
When data science teams are primarily from dominant groups and those perspectives shape the project
What is the privilege hazard?
A company that generates a predictive algorithm that law enforcement uses to gather information to predict/identify potential nefarious actors or predict future events and how to address them
What is Clearview AI?
What two concepts (one from Atlas of AI, and one from Data Feminism) discuss the ability to force a label on someone?
Think classification labels and gender...
What are Power to Define a Person & Binaries?
The classifications that are casually chosen to shape a technical system can play a dynamic role in shaping the social and material world. Thus, we say classification is an act of _____.
What is power?
Life once outside of the “medical field” that is now considered "medical" due to technological advances
What is biomedicalization?
The oxymoron that ignores that information is always situated
What is raw data?
The goal is to identify risks related to the use of the algorithm, with an emphasis on fairness, bias, and discrimination risks. It can involve an external party.
What is algorithm auditing?
A resource that computer scientists and analysts use to create information. It has a context and is "embodied".
What is data?
Overselling AI as being entirely machine-driven, when it's actually powered by humans.
Fauxtomation
The tendency to recast complex social phenomena such as politics, public health, education, and law enforcement as neatly defined problems with definite, calculable solutions or as transparent and self-evident processes that can be easily optimized—if only the right algorithms are in place.
technology solutionism
Which principle of Data Feminism is this? Who does the work? Who benefits? Whose priorities get turned into products?
What is "Examine Power"?
Building a ________is a powerful way of restoring context to data by asking critical questions about a dataset such as where the data came from, who collected it, and why.
What is data biography?
Training data being fed into machine learning without considering the context can lead to this
What is bias?
What is systems of circular logic?
Self-tracking to see how you feel after some activity or event
Eliciting a sensation
"If you refuse to register non-binary people like me ... you do not have the right to .... say that there are not enough of us to warrant change". This quote is relevant to what concept?
What is "what gets counted counts"? or "gender binary"
The kind of machine learning where the process of the algorithm is "learning" from the training dataset.
What is supervised learning?
This is exercised when one group of people uses data in ways that affect other groups of people.
What is power