Skip to content
The Nexus
SCIENCEJun 10 · 17:00 UTCSCIENTIFIC AMERICAN

AI scores a ‘C–’ on its hardest math test yet

The second batch of 'First Proof' problems assesses AI's ability to tackle research-level math. The best-performing model answered six or seven of the ten questions correctly, earning a 'C–' grade.

Nexus surfaces and summarizes. The full story lives at the source.

Mentioned
AItopicFirst Prooftopic
Spot something wrong with this article?Report a problem →
Forward this
Related Signal

Adjacent reporting