# New Study Finds the Structured Output Itself Causes Language Models to Hallucinate

PhantomFill reports that asking thirteen models to fill a form field about an unanswerable input makes them invent an answer they would otherwise decline to give.

- Published: 2026-07-24T05:33:38.818Z
- Canonical: https://polylog.news/ai/2026-07-24/new-study-finds-the-structured-output-itself-causes-language
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [arXiv (cs.LG)](https://arxiv.org/abs/2607.20492)

A paper posted to arXiv, titled "PhantomFill", argues that the form a model is asked to complete, not just the question, causes hallucination. Production language models, the authors note, rarely write free prose. They fill in JavaScript Ob…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-07-24/new-study-finds-the-structured-output-itself-causes-language (subscription information: https://polylog.news/pricing).