> For one, even if a super-efficient FHE scheme was published tomorrow, it'd probably take at least 6-10 years of powerful, sustained cryptanalysis for the comm

> For one, even if a super-efficient FHE scheme was published tomorrow, it'd probably take at least 6-10 years of powerful, sustained cryptanalysis for the community to trust it. Even so, for the use cases suggested (e.g. search engines), it'd still be a vast improvement on what we currently have – even if it later turns out to be flawed.