[Aurornis]

> I don't really get why some people seem to think that it's somehow better to have their bullet point prompt as a huge text.

Simple: It looks like you did more work.

Before everyone had ChatGPT, a long document meant that someone sat at their computer and invested more effort than someone rattling off a list of partially formed bullet points. In the process of writing the doc they usually refined the idea.

Now anyone can dump the bullet points into ChatGPT and get and expand them into a document which gives the illusion of being well thought out. They can now occupy the same space as everyone who was doing a lot of work in the past, but without having to do the work.

[recursive]

That's only going to work until people start absorb the fact that you can now generate unlimited amounts of grammatical text for free. Shouldn't be long now.

[zephen]

Yeah, like yesterday, maybe?

[Terr_]

Next, the recipient of the wall-of-text throws it into another LLM to "summarize" it to bullet-points, adding another hop of the game of telephone as it mutates what was already a fake artifact of thinking that was never done. *sigh*

[filoeleven]

And the money keeps on rolling upward...

[Terr_]

To cynically summarize a disturbing portion of LLM business-models:

1. Find companies which require or reward employees for generating bullshit, fluff, or unreliable signals for actual work that isn't measured well.

2. Offer those employees convenient BS-as-a-Service.

3. Convince the corporation to subsidize the new bullshit-production method as a business expense.

I hope this results in the overloaded collapse of the underlying bullshit mechanics... but I don't particularly expect it.

[imhoguy]

4. Wait for AI token price skyrocket.

5. Approach them to buy your convenient anti-BS-as-a-Service optimizer.

Rinse, repeat.

[SchemaLoad]

But now I can use AI to summarise the wall of text back in to the original bullet points.

[bfivyvysj]

You even get a couple of new ones for free.