The Limitations of AI in Judging Human Behavior
TL;DR
Artificial intelligence (AI) programmed based on rules alone, with no human nuance, is much harsher than when it's based on human responses, according to a study by the Massachusetts Institute of Technology (MIT). The study found that if a descriptive model is used to decide whether a person may re-offend, it may cast more severe judgments than a person and hand out higher bail amounts or longer criminal sentences. The experts have called for greater data transparency, stating if researchers know how data is gathered then they can figure out how it can be used.
Topics:sciencetechnology#artificial-intelligence#data-transparency#human-judgment#machine-learning#mit#technology
- MIT scientists find something alarming about artificial intelligence msnNOW
- Study: AI models fail to reproduce human judgements about rule violations MIT News
- Judging facts, judging norms: Training machine learning models to judge humans requires a modified approach to labeling data Science
- View Full Coverage on Google News
Want the full story? Read the original reporting
Read on msnNOW