Are Large Language Models More Liberal? | Psychology Today Australia - discu.eu

Reddit

Large Language Models appear to be more liberal: A new study of 24 state-of-the-art conversational LLMs, including ChatGPT, shows that today's AI models lean left of center. LLMs show an average score of -30 on a political spectrum, indicating a left-leaning bias. https://www.psychologytoday.com/au/blog/the-digital-self/202408/are-large-language-models-more-liberal 653 comments 1/9/2024 science

Linked pages

The political preferences of LLMs | PLOS ONE https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0306621 0 comments

Related searches:

Search whole site: site:psychologytoday.com

Search title: Are Large Language Models More Liberal? | Psychology Today Australia

See how to search.

Submit link to: