Our readingWhat is a korrent?
Drawn from what Terence Tao said
AI and science Models used to do science rather than to write: rediscovering laws from data, compressing decades of biology, refereeing a proof.
Yeah, this decade I can see it making a conjecture between two things that people would thought was unrelated.
Added to korrents 6 Sept 2026 · How quotes work · Something wrong? Tell us
Do you hold this korrent?Do you also believe this?
I hold this tooIn tension with
Fridman says AI in chess and mathematics explains nothing — it says which position is better and humans build the theory from that. Tao thinks a model could propose a conjecture connecting two things nobody thought were related, with a real chance of being right and meaningful. A conjecture is not an explanation, and it is a good deal more than a score.
In tension with
Current AI models cannot produce insights beyond human understanding, however impressive they seem
Mollick says current models cannot produce insights beyond human understanding however impressive they look. Tao thinks this decade a model could propose a conjecture connecting two things nobody thought were related, with a real chance of being right. Mollick is describing today and Tao the next few years, so the two survive together — but only if something changes in between, and neither says what.
Shared subject first, then similar wording. They may agree or disagree.