Introducing BenchBench

Rohit Krishnan · 25 May 2026

Read the piece on strangeloopcanon.com ↗

Rohit Krishnan did not write this page.

Every claim below was made in this piece, quoted word for word and numbered in the order the piece makes them, so you can read it there rather than take our word for it. The sentence above each quote is our reading of the claim, not their wording. Each quote was checked against a stored copy of the page at build time; where the two differ, the quote is the fact.

Our reading
Designing a good AI benchmark has become a task only the most capable people can do well.

Their wordsCreating benchmarks is now a job relegated to the smartest and best of us.
Our reading
As AI labs aim for broad, general capability, handling messy real-world bureaucratic tasks is becoming a primary application domain for their models.

Their wordsConsidering every lab wants to "eat the world" the focus on how to work in real-world messy situations seems apt as their primary home.

Keep reading

The other pieces are the ones whose claims are worded most like this one's. That measures language, not agreement: they may argue the opposite.