But don't dismiss Go so easily; it hits an interesting sweet spot that may not go away any time soon. It's a simple language with a simple spec, so simple that people are complaining it's too simple a language, yet also simple to use thanks to the GC. But also compiled and fast enough.
But most important of all, it's memory safe and not plagued by undefined behaviour.
Soon (already?) security will mean real money and life or death situation for companies; keeping that much code in a language where nobody can promise a memory corruption will not be introduced in the next commit, is eventually not going to be considered acceptable anymore.
yes, Go is sponsored by a mega corp, yes, and some people cringe at that, but realistically it's less a walled garden than c#, swift or stuff like that.
Rust is likely going to fill the niche currently occupied by C++ but it's quite hard to learn and use.
So yes, it's quite possible that we'll all flock to something new and shiny in 10 years time and forget Go before 20 years have passed. But, whatever replaces Go needs to fill its niche, which if you think about it doesn't have that much free design space left; yes you can improve a few things here and there, but then you have to fight with the massive code base and libraries, that stays relevant due to the absolutely fantastic backward compatibility promises. I've seen C code rot due to compilers getting "better" over time (yes, sure, the C code in question was obviously "wrong", but nobody noticed, because writing correct C code is an exercise in divination)
Rust still has lots to catch up regarding replacing C++ in compiler toolchains, GUI frameworks, game engines middleware and console SDKs, GPGPU, Machine Learning frameworks, HFT, HPC, ....
