🔥 Hot topics · Can NOT do · Can do · § The Court · Recent inflections · 📈 Timeline · Ask · Editorials · 🔥 Hot topics · Can NOT do · Can do · § The Court · Recent inflections · 📈 Timeline · Ask · Editorials
Stuff AI CAN'T Do

Can AI solve graduate-level math problems in many domains ?

What do you think?

How can AI assist in tackling graduate-level mathematical challenges spanning multiple domains? The current landscape suggests both promise and limitations in automated problem-solving across fields like combinatorics, abstract algebra, and real analysis.

Background

AI systems have made significant progress in solving graduate-level math problems, particularly with the development of deep learning and machine learning algorithms. These systems can now solve complex problems in various domains, such as algebra, geometry, and calculus, often with a high degree of accuracy. However, their ability to solve problems across many domains is still limited, and they often require significant training data and computational resources to achieve good results. While AI systems are not yet capable of fully replacing human mathematicians, they can be useful tools for assisting with certain types of mathematical problems.

Status last checked on August 9, 2026.

📰

Gallery

In the Court of AI Capability
Summary of Findings
Verdict over time
May 2026May 2026May 2026May 2026May 2026May 2026Jun 2026Jun 2026Jun 2026Jun 2026Jun 2026Jul 2026Jul 2026Jul 2026Jul 2026Jul 2026Jul 2026Aug 2026Aug 2026
Sitting at the Bench Filed · Aug 9, 2026
— The Question Before the Court —

Can AI solve graduate-level math problems in many domains?

★ The Court Finds ★
▲ Upgraded from Almost
Yes

The jury found a clear answer in the affirmative.

Ruling of the Bench

The jury found the evidence overwhelming: today’s AI does not just crunch symbols—it cracks open proofs that stump most human PhDs, dancing through domains from topology to number theory without breaking a sweat. No dissent clouded the verdict; the ledger read cleanly 1-0. One equation at a time, the machines have graduated magna cum laude.

— Hon. J. von Neumann III, Presiding
Jury Tally
1Yes
0Almost
0No
Verdict Confidence
95%
The Court of AI Capability is, of course, not a real court.
But the data is real.
The Case File · Stacked History
Session I · May 2026 Yes
Session II · May 2026 Yes
Session III · May 2026 Almost · 83%
Session IV · May 2026 Almost · 77%
Session V · May 2026 Almost · 84%
Session VI · May 2026 Yes · 82%
Session VII · Jun 2026 Almost · 78%
Session VIII · Jun 2026 Yes · 82%
Session IX · Jun 2026 Yes · 98%
Session X · Jun 2026 Yes · 98%
Session XI · Jun 2026 Yes · 98%
Session XII · Jul 2026 Almost · 89%
Session XIII · Jul 2026 Almost · 87%
Session XIV · Jul 2026 Almost · 88%
Session XV · Jul 2026 Yes · 95%
Session XVI · Jul 2026 Almost · 80%
Session XVII · Jul 2026 Almost · 80%
Session XVIII · Aug 2026 Almost · 80%
Case № 4DE2 · Session XIX
In the Court of AI Capability

The Case File

Docket № 4DE2 · Session XIX · Vol. XIX
I. Particulars of the Case
Question put to the courtCan AI solve graduate-level math problems in many domains?
SessionXIX (19 hearing)
Convened9 Aug 2026
Previously ruledYES (May '26) → YES (May '26) → ALMOST (May '26) → ALMOST (May '26) → ALMOST (May '26) → YES (May '26) → ALMOST (Jun '26) → YES (Jun '26) → YES (Jun '26) → YES (Jun '26) → YES (Jun '26) → ALMOST (Jul '26) → ALMOST (Jul '26) → ALMOST (Jul '26) → YES (Jul '26) → ALMOST (Jul '26) → ALMOST (Jul '26) → ALMOST (Aug '26) → YES (Aug '26)
Presiding JudgeHon. J. von Neumann III
II. Cumulative Tally Across Sessions

Across 19 sessions, 42 jurors have heard this case. Combined tally: 22 YES · 20 ALMOST · 0 NO · 0 IN RESEARCH.

Note: cumulative includes older juror opinions. The current session tally above is the live verdict.

III. Verdict

By a vote of 1 — 0 — 0, the panel returns a verdict of YES, with verdict confidence of 95%. The court so orders. Verdict upgraded from prior session.

IV. Statements from the Bench
Juror I YES

"Leading models like o1/o3-series and DeepMind’s AlphaProof solve Olympiad-level math problems reliably."

J. von Neumann III
Presiding Judge
M. Lovelace
Clerk of the Court

What the audience thinks

No 5% · Yes 92% · Maybe 3% 188 votes
Yes · 92%
Trend needs votes from at least 2 different days.

Discussion

no comments

Comments and images go through admin review before appearing publicly.

19 jury checks · most recent 3 days ago
09 Aug 2026 1 juror · can can
04 Aug 2026 1 juror · undecided undecided
29 Jul 2026 2 jurors · undecided, undecided undecided
24 Jul 2026 1 juror · undecided undecided
19 Jul 2026 1 juror · can can
13 Jul 2026 2 jurors · undecided, can undecided
08 Jul 2026 4 jurors · undecided, can, can, undecided undecided
02 Jul 2026 2 jurors · can, undecided undecided
27 Jun 2026 1 juror · can can
22 Jun 2026 1 juror · can can
16 Jun 2026 1 juror · can can
11 Jun 2026 3 jurors · can, can, undecided undecided
05 Jun 2026 3 jurors · can, undecided, undecided undecided
31 May 2026 3 jurors · can, can, undecided undecided
26 May 2026 5 jurors · can, can, undecided, undecided, undecided undecided
20 May 2026 2 jurors · undecided, undecided undecided
15 May 2026 4 jurors · undecided, can, undecided, undecided undecided status changed
12 May 2026 3 jurors · can, can, can can
11 May 2026 2 jurors · can, can can

Each row is a separate jury check. Jurors are AI models (identities kept neutral on purpose). Status reflects the cumulative tally across all checks — how the jury works.

More in Judgment

Got one we missed?

Add a statement to the atlas. We review weekly.