AI Nerf: Anthropic's Incident Matches Our Data
Anthropic confirms degraded Claude performance during Aug 29-Sep 4, matching our telemetry data
Hi! Anthropic recently confirmed a performance incident that perfectly aligns with our monitoring data.
Anthropic has published "Model output quality" note confirming periods of degraded responses in Claude models. In particular, they report: "Claude Sonnet 4 requests experienced degraded output quality due to a bug from Aug 5-Sep 4, with the impact increasing from Aug 29-Sep 4". Please see their status page for full details: https://status.anthropic.com


What our telemetry shows:
-
Aug 5–Sep 4: We launched in late August. Even in our short history, results were already jumping around before the Aug 29 spike, and they're steadier after the fix.
-
Aug 29–Sep 4: The issue Anthropic notes is easy to see on our chart - results swing the most in this window, then settle down after the fixes.
Please join our subreddit to stay up to date with the latest testing results: https://www.reddit.com/r/isitnerfed
We're grateful for the community's comments and ideas! We'll keep improving the service for you.