

There are people who get VPNs because they hear that they prevent your ISP from snooping on you when configured correctly, and just hear “no one can see what I do”, because that’s what snooping is, right?
When I worked at a university IT dept, we’d often get content block hits for adult websites from inside the internal protected network, via the university VPN, because a professor or staff member thought a VPN would route their traffic ‘past’ us.
You can’t be sure, but you can use providers and exit nodes that are based in places hostile to whoever you are trying to protect against.
Also, functional anonymity can exist by different entities having different pieces of data that together would de-anonymize you, but who are unlikely to ever intersect. A good example of this is DMCA requests: if a copyright holder sees a US IP address on a residential Comcast IP range, they’re going to file a court case and get a subpoena for the subscriber info.
If they see a Hong Kong IP from a co-lo datacenter who would need to cooperate to tell them who owned that IP at that time, they’re not going to even bother because they don’t know how to even start filing a court case in China, and if your VPN has too much data it won’t even matter because no one will even have contacted them.
It all depends on your threat model.