Machine learning models mirror human prejudices found in large text datasets

Machine learning models mirror human prejudices found in large text datasets
Image: Intellect (generated)

In short: Researchers who applied a purely statistical machine-learning model to a standard corpus of web text found that the resulting artificial intelligence mirrored common human biases. The study replicated various types of associations, ranging from neutral preferences for flowers over insects to problematic stereotypes regarding race and gender.

This research highlights how large datasets serve as cultural archives, capturing not just facts but also historical prejudices. By identifying these patterns, the findings suggest that technology can inadvertently amplify existing societal biases if left unaddressed.

The study used a machine-learning model to analyze text from the World Wide Web. While it demonstrated that these biases are recoverable and accurate, the scope is limited to what exists within the provided corpus.

Why it matters: AI systems can inherit and amplify human prejudices because they learn from the biased patterns embedded in our collective writing.

Opens the Intellect+ feed with this card first. Swipe up for the next one.

Sources

To cite this card, link to https://intellectplus.co.uk/app/science-research-machine-learning-models-mirror-human-prejudices-found-in-lar.

How we source, check and correct cards: Editorial standards.

General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.

Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?

Related

Finding something worth knowing…