Reverse-engineering Claude/GPT knowledge cutoffs and pre-training timelines with daily fact quizzes

Exploring Claude/GPT Knowledge Cutoffs and Pre-Training Timelines

The author built multiple-choice quizzes from daily Wikipedia events to map error-rate curves for GPT-5.4, Opus 4.7, and others. Opus 4.7 onward all share a knowledge cutoff around late December 2025, suggesting a single pre-training base. The GPT-5.6 family comes from a separate checkpoint finishing around late February 2026. Opus 5 is an outlier: its published cutoff is May 2026, but it recalls almost nothing past January 2026—the post doesn't explain why.

Why it matters: The author built a quiz from Wikipedia daily events to map error-rate curves and infer pre-training cutoffs for Anthropic and OpenAI models — clever method, concrete findings. But it's reverse-engineering analysis that appeals more to technical readers, and the excerpt doesn't...