Evidentiality framework labels every model claim as given, verified, or generated — early tests look promising
jzesbaugh · reddit · 2026-10-01
A non-engineer released Evidentiality Framework, a prompt-level provenance scheme: every model claim gets an inline label — (u) given, (m) checked against a named source, (g) generated — plus a parser and gate that blocks actions resting on (g) or unlabelled text. In tests, nine models mostly labelled correctly (some mislabelled their own math as "given"); a five-agent run kept source attribution 33/36 times with labels vs 4/36 without; and a forged "checked" label fooled the next agent 4/4 times, showing labels are an audit trail, not a guarantee.
More from coding & agent
- Early look: Codex Plugin Extensions are cool but still rough around the edges — Angaisb_ · 2026-10-01
- Retriever AI's self-run benchmark scores 24/24 vs OpenAI Dots' 22/24 on five everyday agent tasks — quarkcarbon · 2026-10-01
- Dev builds agent testing tool that catches fake tool calls and loops in full conversations — mbtigeekjung · 2026-10-01
- Dev to agent builders: stop ad channels, kill dark-pattern subscriptions for us — SuB8u · 2026-10-01
- $3,000/month multi-model workflow claims 10-50x productivity gains—and white-collar doom — opmgyhx · 2026-10-01
- "I told my AI tool to replace itself with a cheaper one" — dev celebrates friction-free agent swaps — yacineMTB · 2026-10-01