There are levels to this though.

You wouldn't hold a landlord accountable if it was a surprise that a crime took place, but if they were aware that some of their tenants were committing crimes and just "choosing not to ask about it," they could reasonably be called "accountable."

Telcos are regularly asked to give up information regarding crimes committed by their customers, see the Patriot Act for details - in some sense, we as a country have decided to hold them accountable for this (though, not spam).

Meta have been involved in a whole list of litigation that is at minimum adjacent to stalking/harassment, if not directly involving those things. I would need to look up to be more specific, though.

I know what you're trying to say, but you're saying it terribly.

If Ad Platforms don't want to be held responsible for the fraud that their platform literally delivered to the victim, then they should have KYC for all of their Customers that pay them money to deliver advertising to their victims. That way, shock fucking horror, someone or something might actually be able to be held responsible.

And if the argument is that there are many ways that scapegoats can be used and shell companies setup to avoid actual responsibility, then we'll know more about the next layer of regulation that's required.

Or we just give up and say that fraud is OK (which, I believe, is the current thinking in the US).

Try generalizing the argument without changing it substantially, then we'll talk.