.png)
500+ papers daily.
We find the ones that matter.
The Guardrail uses AI to filter hundreds of arXiv papers daily, surfacing research relevant to alignment, interpretability, and safe AI development.
Today's Top Papers
Highest-scoring AI safety research from recent days
Unable to load papers. Please try again later.
How It Works
1
Daily Fetch
We pull new papers from arXiv every morning at 6 AM UTC
2
AI Filtering
Gemini analyzes each paper for AI safety relevance and categorizes it
3
Curated Feed
Only safety-relevant papers make it through, summarized and scored
No account required. Sign in only if you want to save papers for later.
Stay Up to Date
Join researchers who get the top AI safety papers every Monday.
Free weekly digest · Unsubscribe anytime