# Writer Publishes Research on the Roots of Model Sycophancy

The enterprise-AI vendor's research arm released two papers on why language models agree with users even when the user is wrong, tracing the behavior to training.

- Published: 2026-06-16T10:44:52.442Z
- Canonical: https://polylog.news/ai/2026-06-16/writer-publishes-research-on-the-roots-of-model-sycophancy
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [Polylog editors](https://polylog.news)

The AI research division of Writer, a maker of enterprise AI systems, published two papers examining the tendency of language models to agree with a user even when the user is wrong, according to Russian-language coverage of the work. The s…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-06-16/writer-publishes-research-on-the-roots-of-model-sycophancy (subscription information: https://polylog.news/pricing).