I’m opening the thread to attribute discussion towards the Hidden Inflation Problem. Any ideas previously discussed on the subject would be link-worthy here.
The first question I have regarding this problem: Is there any way for hidden inflation to be detected in a MW blockchain like Grin? My assumption is that there may be way to have a range-proof the UTXO set to show that it doesn’t exceed an expected supply.
That’s exactly what Grin uses. Every utxo has its own rangeproof.
What I meant to say is that maybe there’s a way to sum all the UTXO set and compare that to an expected supply in order to detect hidden inflation. Is this possible?
Block validation does almost exactly that. We sum all the inputs (spent by the block) and sum all the outputs (created by the block) and verify these “sum to zero” by comparing this sum against the sum of all the kernels in the block. We know the expected supply on a per block basis every block has a 60 grin coinbase reward.
Initial “fast sync” does something very similar for the full chain state at the “horizon”, checking that everything “sums to zero” before beginning to download subsequent full blocks. We know the expected supply at the horizon block height as every block has a 60 grin coinbase reward.
The signature on every kernel means every kernel is a 0 value, so all kernels sum to 0 and we need inputs-outputs to also sum to zero for everything to match up.
Note: This is slightly more complex in reality as we introduce a kernel offset per tx, but the core idea still holds, we just need to account for the total offset at the block level.
“Hidden” inflation would cause the sum of inputs-outputs to no longer match the sum of the kernels plus the total kernel offset and validation would fail.
Does that make sense?