🔍 AI Hallucination & Fact Check
🔍 AI Hallucination & Fact Check
Launcher for the standalone AI: extract claims from an answer, check each against your indexed documents, and read coverage, verdict counts and token entropy as hallucination risk. In-browser.
The grounding gauge is a Knowledge-mode instrument: paste any answer, and the standalone AI splits it into claims, verifies each against your documents lexically, flags unsupported numbers and unknown entities, and reports coverage and entropy as risk signals.
Open in the standalone AIOpens at: Knowledge mode → grounding · runs entirely in your browser · no account, no API key, nothing uploaded
What the standalone AI gives you for this
- Per-claim verdicts: supported, partial, unsupported, unsupported-numeric
- Entity check against everything you indexed
- Shannon entropy over the token distribution as a weak surprise proxy
The interactive tool that used to live on this card now runs inside the standalone AI at ai.html, where it shares your indexed documents, stored memories and every reasoning method. This card stays in the catalogue as a launcher so existing links keep working.