According to Friedman and Nissenbaum, a core finding is that…
According to Friedman and Nissenbaum, a core finding is that bias in computer systems is exclusively a problem of poor software implementation or coding errors, and not a problem rooted in the social context, design specifications, or requirements gathering phase.
Read DetailsA major tech company uses a historical hiring dataset, which…
A major tech company uses a historical hiring dataset, which predominantly consists of male employees in technical roles, to train a new AI model for screening and ranking job applicants. When deployed, the AI systematically downgrades applications from women with similar qualifications to men, effectively perpetuating past human hiring biases. According to Friedman and Nissenbaum, what category of bias is the AI demonstrating?
Read DetailsWhen a stochastic parrot LLM generates a racist or sexist re…
When a stochastic parrot LLM generates a racist or sexist response because such content was present in its massive, uncurated web training data, this is best classified as Technical Bias (in the Friedman/Nissenbaum framework) because the underlying flaw is a failure of the algorithm to filter toxic content during the training process.
Read DetailsIf a recommendation engine systematically suggests different…
If a recommendation engine systematically suggests different products to users based on their ethnic surnames, this is a clear example of a biased system only if the discrimination occurs systematically and the outcome (e.g., denying a good opportunity) is judged to be unfair based on unreasonable or inappropriate grounds.
Read Details