I can see how they can open source their client side applications, and I guess they can open source their server code, but I can't wrap my head around how we can verify what exactly they are running on their servers. Like, can we ssh into them with root access and poke around running processes? I just don't get it. Anyone have a clue how else that kind of verification can happen?
I can't vouch for it personally and all I know is that it supports an open-source extension I've relied on for more than a decade, but at least it beats contributing to Karpeles' paycheck, right?
I've heard good things about Mullvad and IPredator, both from Sweden. Generally you want to be looking for VPN providers that don't spend half their revenue on marketing.
I use Mullvad with the standard wireguard clients. Works much better than anything OpenVPN based. They are based in Sweden which seems like a reasonable jurisdiction.