人工智能营销虚假宣传指数
The AI Marketing BS Index

原始链接: https://bastian.rieck.me/blog/2026/bs/

## 人工智能营销 BS 指数 受物理学中“谬误指数”的启发,Bastian Rieck 提出一个“人工智能营销 BS 指数”,用于评估在人工智能产品被过度炒作的时代,营销材料的可信度。该指数旨在识别并评估过于热情或具有误导性的说法。 从 -5 分(给予信任)开始,得分会因以下行为而增加:无正当理由创造术语(+10 分),错误使用科学术语(每次 +10 分),以及使用“这不是 X,而是 Y”之类的修辞手法(+20 分)。 更高分数会授予以下行为:援引“自然”或“宇宙” (+20 分),无根据使用“涌现属性” (+20 分),炫耀常春藤盟校 (+20 分),缺乏可证伪的主张 (+30 分),以及无法验证的研究合作 (+40 分)。 该指数仍在完善中,旨在帮助过滤掉空洞的营销“氛围检测”,并识别真正有价值的人工智能创新。

对不起。
相关文章

原文
The AI Marketing BS Index

You may already be aware of John Baez’s “Crackpot Index”, which provides a simple tongue-in-cheek scoring system for assessing whether a research contribution to physics is likely “revolutionary.” The list contain such gems as “40 points for comparing yourself to Galileo, suggesting that a modern-day Inquisition is hard at work on your case, and so on.”

In the age of large language models, we direly need something similar but for “marketing-speak.” It feels like for every company that wants to actually pitch an interesting product that might do something good for the world, there are at least three others that turned up the “vibe check” to eleven and misuse any bit of terminology they can get their hands on. Thus, a humble attempt at classifying marketing materials along with a grading rubric.

  1. Your concept starts with -5 points. Let’s give everyone the benefit of the doubt.
  2. 10 points for “inventing” something without providing any citation, paper, or other specification.
  3. 10 points for each instance of a term from mathematics, physics, or the life sciences where, by rights, it should not appear.
  4. 20 points for doing the usual motte-and-bailey or hedging in the form of “It is not X. It is Y.”
  5. 20 points for ending a paragraph with pseudo-profound nonsense.
  6. 20 points for claiming that the product does what “nature” does or what “the universe” does.
  7. 20 points for referring to “emergent properties” where this is clearly not warranted.
  8. 20 points for each instance of dropping Ivy League namedropping.
  9. 30 points for having no falsifiable claims or predictions anywhere in the “technical” description.
  10. 40 points for each research “collaboration” that cannot be verified.

As time progresses, I am sure I will have to come up with an extended scoring. Until then, may your inbox be blissfully empty of long e-mails devoid of content.

联系我们 contact @ memedata.com