Anyway, I dont know anyone relevant to me who is using gmail. Most of them are tehnical and they all stopped using gmail years ago, also more and more people are moving to self hosted solutions.
If I'm responding to e-mail from their users, I expect at least my initial response will get through. You really don't need any fancy crap like DKIM/SPF/nice IP addresses database/AI/bayes/etc. in this case to make the decision if e-mail is legitimate.
All my responses contain randomly generated ID of the message I'm responding to that nobody else than the sender and the intended recipient (me) can know. If it matches one of the e-mails their user sent me, say within a reasonable timeframe, and it's a first response or so, it is almost certainly genuine.
It just shows how little they give a crap about their own users, if they don't make such a simple check to make sure that their users can get responses to all sent emails.
Instead of making e-mail work, they're inventing bullshit like this new "expiring email" thing and making redesigns nobody asks for.