Large language models develop novel social biases through adaptive exploration

I don't have access to the full text of this article - only a title and a verification/login page. Based solely on the title, I can tell you that the article appears to discuss how large language models (LLMs) develop new social biases as they adapt and explore, but I cannot provide an accurate summary without access to the actual content. To give you a proper 2-3 sentence summary, I would need the full article text.