Does Wikipedia really need to outsource this? They already do basically everything else in-house, even running their own CDN on bare metal, I'm sure they could spin up an archiver which could be implicitly trusted. Bypassing paywalls would be playing with fire though.