Gosh this is hard. Excellent question. "Existential dread" was a cheap handwave. This is a random collection of thoughts.
First thing I should say is none of what I'm going to say matters if you are the world's most well-adjusted person. i.e. you have 0 inherent desire to do anything other than A) get a paycheck B) do what boss says.
The ex-Googlers I interviewed with put is as "you can't get anything done", which I took as a challenge. And succeeded. Then failed. But not because that's not what leadership wanted, it was the opposite, they had been asking for years: because I was low enough on the totem pole, and in engineering, in a way people could screw around and have 0 consequences.
Hyperbole: every single job I've worked at was better managed than what Google had become at the end, including food service ones. There's a professionalism-as-don't-speak-up authoritarianism in play.
Another handwave I've used to describe it is "humans are humans, everywhere.",
This was the first time I was with a group of really, really smart people. Ivy Leaguers, etc. I dropped out from the #116 ranked university.
I expected that meant the level of shitty behavior would be lower, but, it was higher. Much higher. There's much more training data accumulated in smart people. Making it easier to construct internal narratives for why somethings a bad idea, but, this was usually lazy and slotting into some stereotype. One lesson I learned quickly is there's a parable for every possible decision.
Another problem was you had a ton of people who were just Ivy Leaguers who said "ehhh I guess finance is more evil?", ended up in tech, and really could not fundamentally understand at all why you'd do anything, or talk about anything, or think about anything, other than what someone in authority explicitly said to do and documented and had 0 ambiguity at all. This actually played well with the authoritarianism I mentioned, despite "thriving in ambiguity" being a catchphrase used in perf a lot.
Last thing I'll mention: there were two major decision points that ended up causes an exponential increase in issues:
#1 in...2018?...moving promotion review into the local team, instead of a panel of reviewers from throughout Google. Won't belabor the point too much, TL;DR: that's an excellent way to enable authoritarianism and adjacent issues, because no one can be good enough to just do an end-around.
#2 in late 2022, moving performance reviews from a variety of feedback from collaborators to your manager and whoever your manager bothered picking.
Anyone, if you've made it this far and are very curious, feel free to reach out to jp0hhhh@gmail for deeper conversation (replace the 0 with an o). I'm offering just to pay it forward, i.e. give what I didn't have before Google, someone with a direct sense of tradeoffs.