The scanning is simply going to have to become pervasive.
Manufacturers have used the "spare" weight for aero purposes - deep wheel rims, profiled tubing etc. The incentive to shave off grams is just as strong. Every gram saved by skimping on strength is a gram that can be used to reduce drag.
Safety has been dealt with by the certification scheme for frames and wheels. Every frame and wheelset used in a UCI sanctioned race has to be tested to UCI standards, so the weight is academic.
The technical playing field is far from level and costs have never been higher. The leading WorldTour teams can afford a lot of technology that's out of reach for Continental teams. Two aero bikes, two climbing bikes, a classics bike, a TT bike, eight different wheelsets, plus wind tunnel time to get it all working - that easily adds up to $100k per rider.
Also note that frame efficiency is more important than its weight. For instance there were silly light, silly expensive bikes around which had crap performance, because they twisted like noodles in the wrong places.
Super light weight is a great magnet for dentists.
Weight minimums are good for keeping a level playing field by keeping the cost to entry "low" by keeping it to the thousands.
There are no bikes in the in the World Tour peleton that retail for under $10,000. You can easily build a sub 6.8kg bike for that.
They're also good for safety reasons, both for the rider and others they might crash into; there's no incentive to try to cut too much weight in a way that could impact safety if there's a weight floor anyway.
There's some truth in that, but the limit is way, way lower than where it is at the moment. The fact that weights are being added to push bikes over the limit has no safety impact.
What is the benefit supposed to be of getting hit by a heavier bike?
There's an argument to be made that the technology has improved, and that the limit could be lowered.
[1] A component which has had its strength and durability compromised in the interest of saving weight.