The question is if your diminishing return is less than their diminishing return. My impression is that with fiber connections you have a fairly high cost just because they need to maintain a fiber line, end point equipment, maintenance, service, support, billing and so on. From there they usually offer huge leaps in speed for relatively modest price gains, often like double the speed for 15-20% price gains and that shit multiplies. I could pay about 75% of my current rate to have 20 Mbit instead of 100 Mbit, even though I don't absolutely need 100 Mbit very often it's not worth it. That goes up to a point, then you need some kind of special equipment and the cost skyrockets when you pass out of the "normal" class of equipment and into special gear. Today gigabit isn't actually available to me and if it were it'd cost 200% extra, it's not worth it but if it was 50% I'd probably take it. And my motherboard wouldn't need upgrading.
I'd say 10G is a different story and only about bragging rights at this point, but who knows what the future will bring. If "everybody else" had symmetric gigabit lines, 10G might have a few uses. Sure it costs a bazillion now, but so would a 100 Mbit line not that long ago. It would be a lot more useful to get people on gigabit lines though, it's no good having a huge pipe if nobody can keep up. Already with my 100 Mbit symmetric my upstream is often faster than their downstream, having gigabit would not help at all but if they get upgraded it'd make more sense for me to upgrade. Like for example there's a rural roll-out that'll probably cover my cabin next year, if that's true I could do 100 Mbit offsite, online backup between machines I control. That would be rather neat.