I also think some version of 2 is overwhelmingly likely. In particular, I have always found the core premises of SETI surveys profoundly naive in terms of nece

I also think some version of 2 is overwhelmingly likely. In particular, I have always found the core premises of SETI surveys profoundly naive in terms of necessary preconditions for it to be meaningful. "If your only tool is a giant radio telescope, you look for radio signals." And in specific you look in bands that fit various preconditions of your technology; and then you look for signals which are loosely comparable to the ones we humans used nearly a century ago. The folly of belief that such surveys are likely to uncover unobfuscated communications by some other civilization was truly driven home to me reading here on HN in the last year or two a deep dive into an iPhone exploit that relied on exploiting some issues with its discovery protocols for other devices in the phone's vicinity. The specifics of its multiplexing and frequency hopping and duty cycling across various tasks were hard enough to follow when the priors of the stack and hardware are all very well known. The only signal SETI is liable to see is the equivalent of someone blowing an air horn in a stadium explicitly to be heard. Maybe someone does that. Maybe they find the forest dark enough to not... The more interesting survey IMO is to look carefully for Dyson-sphere style dimming...