斯坦福研究:去年全球英语文稿中24%企业新闻稿、18%消费者投诉内容由AI大模型产出

IT之家
Oct 07

IT之家 10 月 7 日消息,Cell Press 旗下期刊《Patterns》近日刊登斯坦福大学研究团队报告,相应研究团队对 Newswire、PRWeb 和 PRNewswire 等各大平台英语文稿进行分析,发现其中有大量文稿带有 AI 痕迹,认为大模型已被广泛应用于各类正式书面沟通场景。

据介绍,相应团队采用了一套名为“分布式 LLM 量化框架”(Distributional LLM Quantification Framework)的统计模型,从语料整体的语言特征分布中估算特定时间段内含有 AI 生成特征的文字比例。但该方法仅适用于英语,因此上述结果仅反映了英文语料中具备 AI 特征的比例,并不意味着整篇文章“完全由 AI 撰写”。

IT之家参考报告获悉,相应团队分析了去年共计 68.7 万条消费者投诉、53.7 万篇企业新闻稿、3.04 亿份招聘信息以及 1.6 万篇联合国新闻稿,发现其中约 24% 的企业新闻稿、18% 的消费者投诉、14% 的联合国新闻稿,以及接近 10% 的招聘内容“几乎由 AI 大模型直接生成,或经过大幅改写”。

研究人员指出,LLM 已深入渗透社会各领域的正式文本写作,逐渐成为人们日常书面表达的重要辅助工具。尤其是在教育程度较低的地区,使用 AI 撰写投诉的比例达到 19.9%,明显高于教育程度较高地区的 17.4%。在招聘信息方面,中小型企业或初创公司更倾向于使用大模型生成内容,其中约有 10% 至 15% 的招聘公告包含明显的 AI 痕迹。

研究人员同时表示,随着生成式 AI 技术的不断进步,其文风愈发接近人类,未来要精确识别 AI 在文本创作中的参与程度将变得愈发困难。

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Most Discussed

  1. 1
     
     
     
     
  2. 2
     
     
     
     
  3. 3
     
     
     
     
  4. 4
     
     
     
     
  5. 5
     
     
     
     
  6. 6
     
     
     
     
  7. 7
     
     
     
     
  8. 8
     
     
     
     
  9. 9
     
     
     
     
  10. 10