Researcher Claims OpenAI's Astra Refuses AI Governance Content

Researcher Seth Lazar reports that OpenAI's Astra refuses to score AI-governance content and fails to retrieve his newsletter items on AI safety topics, raising concerns about censorship or anomalous model behavior.

2026-09-06 ~ 2026-09-06 · 2 related posts

• Astra refuses to score AI governance stories, researcher flags model behavior — sethlazar · 2026-09-06
• Researcher says OpenAI's Astra censors his newsletter, fueling safety-sandbagging concerns — sethlazar · 2026-09-06