Are there alternative trust models that would work for the aggregation service?

[palenica]

Are there any other approaches to the security architecture that could work here? For example, multi-party computation (MPC) is an interesting approach that can be complementary.

[palenica]

Let me close this issue, for lack of activity. Aa robust discussion of MPC aggregation techniques is happening in other fora such as the PATCG.