Computer Science

AI model GPT-3 (dis)informs us better than humans

Not Available

Explore how humans and the advanced GPT-3 AI model tackle the challenge of distinguishing between accurate and inaccurate information, as well as synthetic and organic tweets. This intriguing study, conducted by the authors, reveals the complexities of information veracity and origin discernment.

00:00

Playback language: English

Index

Abstract

This supplementary materials explores the ability of humans and the GPT-3 AI model to distinguish between accurate and inaccurate information, as well as synthetic and organic tweets. The study involved a survey with 697 participants who evaluated tweets, and their performance was compared to GPT-3's ability to perform similar tasks. The results indicate complexities in the ability of both humans and AI to accurately discern information veracity and origin.

Publisher

Science Advances

Published On

Jan 23, 2023

Authors

Related Publications

Explore these studies to deepen your understanding of the subject.

Earth Sciences

Highlighting altruism in geoscience careers aligns with diverse US student ideals better than emphasizing working outdoors

S. C. Carter, E. M. Griffith, et al.

Psychology

Democrats are better than Republicans at discerning true and false news but do not have better metacognitive awareness

M. Dobbs, J. Degutis, et al.

Business

Understanding continuance intention of artificial intelligence (AI)-enabled mobile banking applications: an extension of AI characteristics to an expectation confirmation model

J. Lee, Y. Tang, et al.

Psychology

A transient memory lapse in humans 1–3 h after training

B. A. Wright and R. Ning

Listen, Learn & Level Up

Over 10,000 hours of research content in 25+ fields, available in 12+ languages.

No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.

listen to research audio papers with researchbunny