Testing Fable: Two Factual Hallucinations in 10 Minutes

mattpocockuk · x · 2026-07-06

Renowned TypeScript educator Matt Pocock reported encountering two factual hallucinations within just ten minutes of using Anthropic's new model, Fable, warning against blindly trusting LLM outputs. This observation highlights that the new flagship model still falls short in factual accuracy.

Related event: Matt Pocock Says Fable Still Hallucinates(2 posts)→

Original post →

More from Models

Models channel →