4 ms·

Researchers Asked LLMs for Strategic Advice. They Got "Trendslop" in Return

17 points by johnbarron 5mo ago

johnbarron 5mo ago
"Harvard Just Caught AI Lying to Every Executive in America" - https://youtu.be/pd1Km6bT104 https://youtu.be/pd1Km6bT104
- jdlshore 5mo ago
  I watched the first half of this video before giving up. It’s sensationalized and I’m not convinced the YouTuber actually knows what he’s talking about. You’re better off reading the actual article, which is very good.
  - johnbarron 5mo ago
    Mostly agree with everything you said. I hesitated before posting the link, but the video gets better in the second half of it... :-), and though would be a second change for people, who would fail to engage with the high quality article.
bediger4000 5mo ago
Well of course they did! "AI" can't possibly replace Accenture or McKinsey, or lawyers!
tim-tday 5mo ago
What else would they expect? Next token prediction virtual requires this outcome.
Selkirk 5mo ago
I'm missing some data to evaluate this article. If you engage a set of humans on these seven tensions, what range of bias do you get?
cpharsh410 5mo ago
[dead]
jdlshore 5mo ago
Lots of middlebrow dismissal in the comments on this article, but none that actually engage with its content, at least not at the time I’m writing. Some of the dismissals seem like they only read the headline, not the article. (Shocking, I know.) Personally, I found the article well written, and describing a very thorough set of research. I don’t doubt that there are leaders right now ceding their decision-making to LLMs, and they would do well to read this article.
CamperBob2 5mo ago
>An LLM is not the colleague who critically evaluates current ideas, looks into the contextual specifics, stress-tests assumptions, and pushes back when everyone gets comfortable Well, not if you don't ask it to, it won't. So ask it to. (The article, which is better than it sounds like it's going to be at first, actually does advocate that towards the end. It's worth a read.)

17 points by johnbarron 5mo ago

"Harvard Just Caught AI Lying to Every Executive in America" - https://youtu.be/pd1Km6bT104 https://youtu.be/pd1Km6bT104

jdlshore 5mo ago
I watched the first half of this video before giving up. It’s sensationalized and I’m not convinced the YouTuber actually knows what he’s talking about. You’re better off reading the actual article, which is very good.
- johnbarron 5mo ago
  Mostly agree with everything you said. I hesitated before posting the link, but the video gets better in the second half of it... :-), and though would be a second change for people, who would fail to engage with the high quality article.

bediger4000 5mo ago

Well of course they did! "AI" can't possibly replace Accenture or McKinsey, or lawyers!

tim-tday 5mo ago

What else would they expect? Next token prediction virtual requires this outcome.

Selkirk 5mo ago

I'm missing some data to evaluate this article. If you engage a set of humans on these seven tensions, what range of bias do you get?

cpharsh410 5mo ago

[dead]

jdlshore 5mo ago

Lots of middlebrow dismissal in the comments on this article, but none that actually engage with its content, at least not at the time I’m writing. Some of the dismissals seem like they only read the headline, not the article. (Shocking, I know.) Personally, I found the article well written, and describing a very thorough set of research. I don’t doubt that there are leaders right now ceding their decision-making to LLMs, and they would do well to read this article.

CamperBob2 5mo ago

>An LLM is not the colleague who critically evaluates current ideas, looks into the contextual specifics, stress-tests assumptions, and pushes back when everyone gets comfortable Well, not if you don't ask it to, it won't. So ask it to. (The article, which is better than it sounds like it's going to be at first, actually does advocate that towards the end. It's worth a read.)