diff --git a/hive-matrix-mcp/src/main.rs b/hive-matrix-mcp/src/main.rs
index 6fb90b35..40c78aab 100644
--- a/hive-matrix-mcp/src/main.rs
+++ b/hive-matrix-mcp/src/main.rs
@@ -85,16 +85,8 @@ const ACCOUNTS_CHANGED_EXIT: u8 = 75;
 async fn main() -> Result<ExitCode> {
     tracing_subscriber::fmt()
         .with_env_filter(
-            tracing_subscriber::EnvFilter::try_from_default_env().unwrap_or_else(|_| {
-                // matrix-sdk-crypto's own WARN fires every sync tick on any
-                // account with no key-backup set up (the common case here) —
-                // measured at ~76k lines/day swarm-wide, the single largest
-                // log message by count. The unit normally sets RUST_LOG
-                // explicitly (see matrix.nix), so this fallback mainly
-                // covers a bare `cargo run`; an operator's explicit
-                // override still sees it if they ask.
-                tracing_subscriber::EnvFilter::new("info,matrix_sdk_crypto::backups=error")
-            }),
+            tracing_subscriber::EnvFilter::try_from_default_env()
+                .unwrap_or_else(|_| tracing_subscriber::EnvFilter::new("info")),
         )
         .with_writer(std::io::stderr)
         // This is a systemd-managed daemon — stderr always goes to journald,