Blog

Triadic-dependent and collaborative-dependent models of consultation are helpful when seeking change for

“The __________ model would more effectively address complex problems, like the multi-causal and multi-contextual problems of at-risk youth. “

__________ may enhance the triadic dependent consultation process.

In the triadic dependent model

In the traditional consulting relationship

The paper argues that when language models are created using training data collected without adequate documentation (data sheets) about its composition, potential biases, and filtering methods, this leads to an ethical risk they call “Data Amplification,” where errors are replicated across the web.

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.

An automated legal expert system (AI) designed for the U.K. is initially trained and tested exclusively using established U.K. law (statutes and precedents). Over time, a major international treaty—which significantly alters the interpretation of citizenship rights—is ratified, but the AI’s knowledge base is not updated. The AI begins to systematically misinform individuals whose rights are defined by the new treaty. This scenario best illustrates which category of bias?  

Bender et al. argue that the high environmental and financial costs associated with training ever-larger models are largely justified, as the accompanying advancements in safety and utility outweigh the resource expenditure.

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?