You can't ever win this argument against the public and the powers that be by staking it all on "but privacy is important, don't you like privacy?".
That's what they keep reducing it to. They're also making it a false dichotomy of sorts, but in reality it's a gradient of possibilities. For example, VPNs aren't like Tor in that they can't really resist "NSA" level global wiretap monitoring in any meaningful way. Or even ISP-level data-analysis driven investigations.
It's also important to correlate any privacy protections VPNs provide, with a real-world pre-internet equivalent. paper mail for example has always been subject to Mitm by the authorities. It is possible to divulge who visited what site, and at what time, and only directly to the authorities, and make that disclosure public (after gag orders expire, if any are issued).
You can use VPNs for privacy against all sorts of creepy eyes, but your local government being considered one of those hostile actors is the threat model that's under attack here.
I would argue for example that the pre-internet equivalent would be two people chatting in the privacy of one of their homes. A bit of a stretch, but alright. But in that there must be the element that the two persons are able to identify each other positively. If one of them is harmed by the other, the victim can identify the attacker to the authorities and pursue justice. How can that be done with VPNs? If middle-actors can't snoop, then can logs on both ends positively identify the other party? Was there a common way pre-internet, where people anonymously gathered and discussed things, with capability to harm each other, but without the authorities being able to do anything about it after the fact?
If the authorities are able to gain access to a private key, or some other proof of possession of one end of the connection, can the VPN provider, the network, or the protocol disclose the identity of the source of traffic on the other end?
I'm only making these arguments to point out how nuanced the topic is. The false dichotomy of all-or-nothing for VPNs is silly. this is moving towards an outright ban of VPNs with criminal consequence, and with that all other similar tech (including Tor) and privacy measures go down the toilet. Would you rather have that or propose a nuanced compromise one jurisdiction at a time?
I get this is just PR for Mozilla though.