Here's another testable area in which I am almost certainly wrong, but I think a real investigation is possible:
Bracewell Probes. We can posit a reasonable concept of operations for a Bracewell Probe, try to figure out what the observables might be, and then go looking for them. I offer the hope that there are already astronomical data out there that may contain confirming evidence. This needs more development, but I think that conducting the search would be highly instructive.
I agree.
The problem with searching for advanced probes that I see are:
1. they might be small. Like really, really small. In 100-1000 years we might be making such probes the size of baseball or golf balls.
2. they might not want to be found.
3. we might not recognize them as technology at all.
4. if they don't communicate via radio waves, we might miss them.
Personally I'd look for odd orbitals, and certainly things that shift orbits. High albido, but again they'd have to want to be found.
Higher thermal output that can be accounted for via blackbody radiation is another one. Like it or lump it, but RTGs are damn efficient, easy to build, and simple -- meaning they could conceivably last thousands of years (Am for example could last 1000 years), with the added plus of keeping your probe from freezing.
If I were to send such a thing to a solar system that might contain life, I'd send a few thousand in a carrier probe, and then shotgun them upon insertion to the system. They could communicate short-range with the carrier, which then could tight-beam laser back home.
Thousands would make the odds of success high, and you wouldn't sweat losing 99% of them.
If they were small you could use ion propulsion or something to let them scatter about the target system seeking life or whatever else you were looking for.