New research into large language models shows that they repeat conspiracy theories, harmful stereotypes, and other forms of misinformation.

In a published study, researchers at the University of Waterloo systematically tested an early version of ChatGPT’s understanding of statements in six categories: facts, conspiracies, controversies, misconceptions, stereotypes, and fiction. This was part of Waterloo researchers’ efforts to investigate human-technology interactions and explore how to mitigate risks.

They discovered that GPT-3 frequently made mistakes, contradicted itself within the course of a single answer, and repeated harmful misinformation.

Though the study commenced shortly before ChatGPT was released, the researchers emphasize the continuing relevance of this research. “Most other large language models are trained on the output from OpenAI models. There’s a lot of weird recycling going on that makes all these models repeat these problems we found in our study,” says Dan Brown, a professor at Waterloo’s David R. Cheriton School of Computer Science, who co-authored the work with Aisha Khatun, a master’s student in computer science.

“There’s no question that large language models not being able to separate truth from fiction is going to be the basic question of trust in these systems for a long time to come,” Brown says.

From University of Waterloo
View Full Article