AI scores a ‘C–’ on its hardest math test yet
The second batch of “First Proof” problems is meant to evaluate AI’s usefulness for research-level math. The best model got six or seven of the 10 quest…
SC
Scientific American
10 Jun 2026 8 days ago 1 min read
Scientific American — 10 June 2026
Text:
3 0 0
🌐 Translate:
🎙️ AI Podcast — Two-Host Discussion
AI scores a ‘C–’ on its hardest math test yet
Kokoro TTS · ~5 min episode · American English voices
Choose voices for Host A and Host B. Changes take effect on next play.
A
Now Speaking
—
Generating podcast…
Synthesising speech — will play automatically when ready…
0:00 / 0:00
The second batch of “First Proof” problems is meant to evaluate AI’s usefulness for research-level math. The best model got six or seven of the 10 que