I heard that Spectre could be used to reveal arbitrary information from RAM and I imagine it should ultimately be possible to automate getting root this way. At the same time, quite a couple of months passed since this class of vulnerabilities was disclosed and I still hadn't seen any convincing exploit.
Normally I would assume that this is in order to keep script kiddies out, but given that we're likely to stick with broken processors for a few years, anybody who would publish such program would definitely get popular.
This leads me to the question: is it actually possible to get local root via spectre? If so, what could be the reasons why no exploit was published yet?