And I'm wondering, are we even always talking about humans that should spot these things? Or is it also a lot of like automated monitoring that takes place?

Indeed.

So PayPal relies on algorithms to kind of detect that its merchants are doing the right thing in a certain sense.

However, my sources tell me that these algorithms are built to catch fraud.

They're not built to scan for extremism.

My perception from reporting is that there's an idea that the algorithms will catch extremism or will catch things that are completely illegal or so on and so forth.

But in practice, it does not seem like that is being very effectively done.

And at the same time, advocating from sort of a devil's advocate perspective, you could also say, well, isn't this free speech? Shouldn't like anyone be allowed to make these things and sell them through PayPal? How would you look at that argument?