I've seen some rather atrocious techniques employed to bake ads into source but that can still be done without any performance impact. Making that work in WASM would require dynamic recompilation, doable perhaps for smaller apps but for behemoths like Gmail and Docs I couldn't see even Google getting that to scale. I think we'll still have browsers performing additional HTTP requests for that stuff, at least in the short to medium term. So traditional adblocking techniques will still be viable.

It will be like cutting out the ads in a newspaper. Lots of holes, everywhere.

The difference from a physical newspaper is that on a computer we don't have to leave holes, we can replace them with beautiful artwork or our favorite cat pictures. That's still a hell of a lot better than ads.

I'm ok with ads in a newspaper. I can ignore them safely and blend them out. Many online ads nowaday blink or do other dark patterns to make ignoring them harder though. I hate that.

That still defeats the purpose. I want the ads to be blocked from loading for performance reasons more than anything else.

>run an AI model

what exactly would this AI model be trained on?

Ads? The suggested approach is "from pixels" so any rendered ad is a source of training pixels.

Is there any ad blocker that blocks ads rendered within videos?

It's possible doesn't mean it's viable.

SponsorBlock does a surprisingly good job: https://sponsor.ajay.app/