Scholarlink

<p>It has been previously argued (Hicks et al., <CitationRef CitationID="CR7">2024</CitationRef>) that LLMs should be described as <i>bullshitting</i> – rather than “hallucinating” – because of how they generate text. Licon (<CitationRef CitationID="CR10">2025</CitationRef>) offers a “complementary explanation” for the dissemination of bullshit. He suggests that they produce bullshit because of the prevalence of bullshit in the training data. We contend that this is mistaken. While ChatGPT does produce bullshit, this is because of its process.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

LLMs Bullshit by Design: A Reply to Licon

  • 期刊论文

  • 发表日期：

    2026年5月11日

查看全文
  • James Humphries,
  • Michael Townsen Hicks,
  • Joe Slater

摘要

It has been previously argued (Hicks et al., 2024) that LLMs should be described as bullshitting – rather than “hallucinating” – because of how they generate text. Licon (2025) offers a “complementary explanation” for the dissemination of bullshit. He suggests that they produce bullshit because of the prevalence of bullshit in the training data. We contend that this is mistaken. While ChatGPT does produce bullshit, this is because of its process.