I think a lot of that is people who were always very insecure because they’re mediocre engineers. Previously asking questions or not understanding something was a bit painful but now you can hear “you’re absolutely right!” and make progress all day every day. Until you need to interact with actual humans and that’s why you try to minimise it, hiding behind ai generated content.

Everyone I know, including myself, is simultaneously extremely pro-AI and anti-AI for this reason.

I always put myself in the shoes of the reader when relaying Claudeslop. I either use it as a reference and write it by hand or make Claude edit it 10+ times until it's conveying only the needed information, and in simple, clear words, and without annoying Claudeisms.

He's basically asking other people to do his job for him, because he knows that the AI doesn't truly do it well enough.

I would say to him that if he wants me to do his job, he needs to pay me his salary.

I am that person on the blast radius. The most effective response I have found is dumping the documents into the AI, and asking it to review them, then posting back the AI response.

Any resistance to this is met with a "But I'm also using AI as the board demands!"

> Au psychosis is getting seriously outrageous at this point

So... back to a currency backed by gold?

How correct is the documentation? You seem to be coming from the same standpoint as a lot of people in this thread which is "don't make me read AI output, I can prompt the model myself" but what this person is doing sounds much more sane - have highly skilled subject matter experts prompt the AI once, generate all the answers you might want, and review them carefully for correctness using all available context.

That sounds far preferable to individuals independently asking questions that are not covered by the documentation and independently reviewing a bunch of potentially hallucinated slop that no one else will ever read.

On the contrary, I kind of love this as form of malicious compliance. I'd wager money that the PO/BAs have been demanding "comprehensive documentation" for a very long time, and now they are getting what they've asked for good and hard.

AI be used to make reviewing it easy, eh?

It's okay, when enough terrible decisions impact a company they tend to do really poorly thought out layoffs so you'll be in the blast radius sooner or later.

Too bad there's no real accountability.

>Thankfully I'm a dev and thus aren't in the blast radius of that genius idea

yes you are

When some people call this "Verification Debt" or something similar, that is setting the Overton Window upon this phenomena adversely upon the verifiers. I call it an application of Brandolini's Law:

"The amount of energy needed to refute bullshit is an order of magnitude bigger than that needed to produce it.""

Until you put in at least the same amount of energy required to verify the AI output as you are asking others to put in to verify it, the AI output you are wielding must be considered bullshit. Really, it should be based upon the verification requestor's historical trust score and how many human verifiers are looped in. A Verification Tax, so to speak:

The energy the creator must spend (Ec) is determined by a function of the number of reviewers requested (N) and the historical trust score of that creator's AI outputs (T).

Ec = N * (1 - T)