>“There’s no way that you can do a full investigation of every possible signal that you detect, because doing that still requires a person and eyeballs,” he said. “We have to do a better job of measuring what we’re excluding. Are we throwing out the baby with the bath water? I don’t think we know for most SETI searches, and that is really a lesson for SETI searches everywhere.”
Is this not the perfect job for AI today? Just sit there and digest signals for 30 years and report back the top 1000? I'm quite sure it could even work on the algorithms as a side-quest.
Claude Code: i'm entering plan mode to analyze the 10B signals in the database
Digest signals for 30 years and report back? That's one hell of a super computer and significantly faster than Deep Thought
No, AI are terrible at finding these types of patterns.
You could hypothetically use AI to write algorithms to find the patterns, but people have already spent a long time super-tuning them.
AIs can't even (at least I keep checking) solve Sudokus as well as my mother -- they aren't good with piles of numbers and complex patterns.
AI has been used in astrophysics for a long time, now. AI is more than genAI of the past 3-4 years... Classification tasks are handled by AI now because it's the only thing that gives you some accuracy at scale.
If nothing else, AI will probably be needed to filter out RF artifacts and spurious emissions from all the Internet satellite constellations that are either already online or ramping up in the future.
This sort of effort really ought to be conducted with antennas on the far side of the Moon, IMO. But good luck finding the budget for that these days.
