The cognitive radar stuff is old tech. I don’t think that concept is really considered a differentiated capability beyond being a sophisticated implementation.
Almost by definition, any classified computer science research would be non-obvious.
If cr is old tech any keywords for what is new/current tech?
I’m not sure your second point is true. The vast vast majority of classified information is very boring, or operational like frequencies of radar, etc.
Both sides know the basics, it’s what frequencies the radar comms and aircraft work at that is classified.
There’s very little “OMG this one algorithm changes everything!!”. Unless proven otherwise
That is entirely the point, it is supposed to be surprising. There are fragments of circumstantial evidence for some interesting computer science problems e.g. systems that demonstrably imply transitive closure algorithm performance that can’t be remotely replicated by anything in literature.
The ability of someone to imagine the existence of things they are unaware of has no bearing on their existence. You could say the same of a lot of the classified materials science that underpins a lot of US weaponry hardware for which there is ample circumstantial evidence. No one is going to be talking about it on HN.
Almost by definition, any classified computer science research would be non-obvious.