You don't get it: I don't want any dependencies, period. I don't want a build step, period.
And even if I do have some dependencies and a build step, the last thing I want is to install 30 more only to deal with the ungodly mess that Tailwind generates. Now not only I have to write my styles in an arbirary new Tailwind-specific syntax that in some cases resembles nothing like CSS and more like Brainfuck, but I also have to keep ## packages updated just to use a new CSS feature that my current Tailwind version doesn't support. To me it's flat-out insane. Not even having a baseline design token is a winning point for me, because for that I could use something much lighter like Open Props that doesn't impose a new organizational colonization of my codebase.
The only case that Tailwind is barely sustainable for me is for quick scrap html pages that I could throw in the bin right afterwards, and even in that case there are better alternatives. In all other cases, I have ZERO incentives to use a library like Tailwind, which not only is a disaster, but also disastrous in the sense that's actively making a mess out of codebases, and making developers think they can style while they're actually doing worse than those you couldn't add two numbers together without jQuery (back in the day) or React.
I read around of people asking how to solve a styling problem with Tailwind without realizing that the same problem has been solved by plain CSS for decades with what should be considered common knowledge. So no, I don't fall for the usual "If you like it, use it; if you don't like it, don't use it", and I'm taking a much more active role in this and try to convince everyone that Tailwind and similar atomic/utility CSS frameworks are a terrible idea all around.
The title of the video you linked should be "Tailwind is the worst." PERIOD, no ellipsis, because all the point he's trying to make are easily refutable and nowhere near to be compelling arguments. So no, don't think for a single second that you've brought valid arguments. Basically everyone that I know that actually know CSS, talk about CSS (as conference speakers), are passionate about styling and even have a role in shaping CSS' evolution, all have a very low consideration of Adam Wathan's sales pitch. And that's for a reason (multiple reasons, actually).