New accounts on HN more likely to use em-dashes (marginalia.nu)

Article summary

An analysis of comments on Hacker News found that new accounts are nearly 10 times more likely to use em-dashes and other symbols, and also more likely to mention AI and LLMs. The author suggests that this may indicate an influx of bots on the site. The analysis was based on a sample of around 700 comments from new and established accounts. The findings have sparked a discussion about the presence of bots on the site and how to distinguish between human and AI-generated content.

Main themes

• bot detection
• AI-generated content
• online discourse
• authenticity
• homogenization of opinions
• bot prevention strategies

What commenters say

• Some commenters believe that the presence of bots on the site is a significant problem that needs to be addressed.
• Others argue that it is difficult to distinguish between human and AI-generated content, and that any attempts to do so may be gamed by bot operators.
• There is a concern that the use of AI to generate comments may lead to a lack of diversity in writing and a homogenization of opinions.
• Some commenters suggest that a possible solution to the problem of bots is to implement an invitation system or to shade accounts that are less than a year old.
• Others propose that users should be able to see how much of the engagement they are seeing is from bots.
• It is also argued that people should not use AI to generate comments, as it can lead to a lack of authenticity and a decrease in the quality of discussions.
• On the other hand, some commenters believe that AI-generated comments can be useful and that they should not be automatically dismissed as spam.
• There is also a concern that any attempts to restrict or label AI-generated content may be met with backlash from users who rely on AI to write their comments.