If I were one of the Wright brothers watching this thread I suspect that I would not join the experts in posting speculation on whether passengers could or could not be released from a fastened seat belt by g forces of unknown magnitude. Instead, I might go to my bicycle workshop and fix a seat belt (if I had one) to the outside of a bicycle wheel with the buckle release lever outwards away from the center of the wheel. I migght then spin the wheel at gradually increasing speed until the centrifugal action eventually caused the release lever to rotate thereby releasing the buckle. From the rpm and the radius of the wheel I would then know the g force needed to release the buckle. In one hour I would have the answer.
You could do this. And if so I would be interested to learn the answer. Until then, here is my guess (and it is just a guess). The buckle flap won't simply fly open, at least not until you reach some fantastic rpm. The flap is not that massive and won't be very susceptible to this centrifugal force. Depending on design, it may not be at all susceptible. But it won't be susceptible enough for this to happen at the sort of force we are talking about. What you could measure would be increased tension along the belt as the circle (ie the loop of belt and buckle) sought to enlarge - but this tension is what the belt is designed to oppose.
I realize that some QF72 pax claimed their belts failed. And just maybe they did get caught on something. But I also know that Qantas routinely advises pax to wear belts even in smooth cruise. And so any pax seeking damages would have thought that they better have been wearing a fastened belt.