The title of the video is unfortunately named (because it's a click-baity title), but the content ISN'T wrong.
The way it is presented in this thread, without context is the issue. He talks about not all random feedback is good feedback, and establishes the proper conditions to obtain feedback. At
10:34, he says in regards to proper feedback:
"I think it should be done in a very focused, clear controlled environment. I like the idea of you getting 95% of the way there, with a particular idea - not the whole product - but 95% of the way, with like a chapter of your book or an idea of your story and you go... ah fuck, but I'm wondering if it's a little too confusing at this part or whether it makes sense. You hand it to someone and go - give me your reaction... "
Does that sound familiar? This Early Access process by Larian is exactly this lol.
He does make other points about what makes useful feedback - which is also fair IMO. We've seen people make feedback based on faulty information that should be ignored - e.g. we've seen people complaining that BG3 melee is weak without realizing there is a backstab mechanic.
BG3 is an extra interesting scenario because of the game's attachment to both the established 5E D&D ruleset and the BG1/2 systems - so now you're dealing with up to 3 conflicting visions for BG3 - (the Larian Game, the BG1/2 Sequel, and the 5e Videogame). I'm in the camp that Larian should make the game they want to make, but I know there are people that vehemently disagree with that.