AWS is fine: https://health.aws.amazon.com/health/status

My guess is whatever system downdetector uses to "detect downtime" relies on either GCP or Cloudflare (also having issues at the moment: https://www.cloudflarestatus.com/)