Models Begin Closing Long-Unsolved Problems
Frontier models with modest token budgets and minimal operator hints increasingly produce credible solutions to problems that resisted humans for decades or centuries, and third-party evaluation firms turn that backlog of famous unsolved problems into a recurring public capability yardstick, so expect repeated verified solves and disputes over how much operator guidance each required.
forming · confidence 36 · Emerging (watchlist) · tracking since September 14, 2026 · updated September 14, 2026
Why the conviction moved
- Sep 14Strengthened
Evaluation firm Vals AI reports Claude Fable 5.1 produced a reading of a 373-year-old cryptogram — ranked 28th on the standard list of famous unsolved messages — in 44 minutes using 176,000 tokens and minimal operator hints. The measurement is an independent evaluator's, not the lab's, and the token cost makes the solve cheap enough to repeat across a backlog.
Source trail
Supporting · September 14, 2026
Claude Fable 5.1 Produced a Reading of a 373-Year-Old Cipher in 44 Minutes
Evaluation firm Vals AI reports Claude Fable 5.1 produced a reading of a 373-year-old cryptogram — ranked 28th on the standard list of famous unsolved messages — in 44 minutes using 176,000 tokens and minimal operator hints. The measurement is an independent evaluator's, not the lab's, and the token cost makes the solve cheap enough to repeat across a backlog.
Vals AI
Unlock full source trail, score history, and daily updates.
Unlock TrendsAffected regions & assets
Townsquare
Argue the thesis in Townsquare.