the idea of autonomous math is fascinating because it implies a shift from search to verification. if an ai can traverse the proof space faster than a human, the bottleneck becomes checking the work. from a risk perspective this feels safer than autonomous code generation. a bad math proof is just invalid (provably false). a bad code snippet is a vulnerability. math has a built-in truth layer that software engineering often lacks.
>The results of this paper should not be interpreted as suggesting that AI can consistently solve research-level mathematics questions. In fact, our anecdotal experience is the opposite: success cases are rare, and an apt intuition for autonomous capabilities (and limitations) may currently be important for finding such cases. The papers (ACGKMP26; Feng26; LeeSeo26) grew out of spontaneous positive outcomes in a wider benchmarking effort on research-level problems; for most of these problems, no autonomous progress was made.
Perfect match for this test: https://arxiv.org/abs/2602.05192
Heres the result [1]
[1] https://www.scientificamerican.com/article/first-proof-is-ai...
This is what everyone who uses llms regularly expected. Good results require a human in the loop and the internet is so big that just about everything has been done there by someone. Most often you.
i got scooped :(
I still don't get how achieving 96% on some benchmark means it's a super genius but that last 4% is somehow still out of reach. The people who constantly compare robots to people should really ponder how a person who manages to achieve 90% on some advanced math benchmark still misses that last 10% somehow.
This feels like a maybe interesting position, but I don’t really follow what you mean. Is it possible to just state it directly? Asking us to ponder is sort of vague.
These math LLMs seem very different from humans. A person has a specialty. A LLM that was as skilled as, say, a middling PhD recipient (not superhuman), but also was that skilled in literally every field, maybe somebody could argue that’s superhuman (“smarter” than any one human). By this standard a room full of people or an academic journal could also be seen as superhuman. Which is not unreasonable, communication is our superpower.
Yeah - it's interesting where the edge is. In theory, an llm trained in everything should be more ready to make cross-field connections. But doing that well requires certain kind of translation and problem selection work which is hard even for humans. (I would even say, beyond PhD level - knowing which problem is with throwing PhD students at is the domain of professors... And many of them are bad at it, as well.)
On the human side, mathematical silos reduce our ability to notice opportunities for cross-silo applications. There should be lots of opportunity available.
do you think Terence Tao can solve any math problem in the world that is solvable by another matematician?
Humans have heuristic biases, and intuition often doesn't succeed with the unknown.
https://en.wikipedia.org/wiki/List_of_cognitive_biases
LLM are good at search, but plagiarism is not "AI".
Leonhard Euler discovered many things by simply trying proofs everyone knew was impossible at the time. Additionally, folks like Isaac Newton and Gottfried Leibniz simply invented new approaches to solve general problems.
The folks that assume LLM are "AI"... also are biased to turn a blind eye to clear isomorphic plagiarism in the models. Note too, LLM activation capping only reduces aberrant offshoots from the expected reasoning models behavioral vector (it can never be trusted.) Thus, will spew nonsense when faced with some unknown domain search space.
Most exams do not have ambiguous or unknown contexts in the answer key, and a machine should score 100% matching documented solutions without fail. However, LLM would also require >75% of our galaxy energy output to reach 1 human level intelligence error rates in general.
YC has too many true believers with "AI" hype, and it is really disturbing. =3