What Grant Sanderson thinks about benchmarks

@grant-sanderson · 25 positions · On ppll ↗

Mathematics explainer, creator of the YouTube channel 3Blue1Brown.

Grant Sanderson did not write this page.

We collected these quotes from things they published elsewhere, and every quote links to where it was said. They have no account here and have not endorsed this site. Quotes are word for word; the short line under each one is our own restatement, not their wording. Their own site. Is this you? Claim it or ask us to remove it. Or tell us what is wrong here.

Filter benchmarks · Newest first

Topic

All 25mathematics 14creativity 2AGI 1AI agents 1AI slop 1AI writing 1benchmarks 1education 1formal proof 1

4 more topics

Google 1LLMs 1reinforcement learning 1writing 1

Sort

NewestOldest

Only Grant Sanderson on benchmarks. See everyone on benchmarks

1 dated position, 2026, in their own words. Our reading of what Grant Sanderson has said — not written or endorsed by them.

1 position so far — this page is not yet offered to search engines.

1. 30 Jun 2026

   Our readingAI learning to generate good conjectures will never show up as a benchmark being knocked down; it will show up as a shift in how mathematicians talk about the tools.

   mathematics

   Their wordsI think the way that you'd measure conjecture generating ability is going to be more subjective on like that tone shift where um it'll be mathematicians saying they're not just using it to like solve their problems, but as they step back and decide what their research field should even be that a conversation with such and such model like was genuinely helpful for that.

   ↗Grant Sanderson (@3blue1brown) – AI disproved a famous math conjecture. Now what?youtube.com 2nd of 23 in this recording