Language models (aka LLMs) are known to produce overconfident, plausible falsehoods, which diminish their utility and trustworthiness. This error mode is known as “hallucination,” though it differs fundamentally from the human perceptual experience. Despite significant progress, hallucinations continue to plague the field, and are still present in the latest models.

An interesting read.

https://arxiv.org/abs/2509.04664

First dropped: March 8, 2026 | Last modified: March 08, 2026