Our test that watches for alerts to appear we've never seen before has picked something up on https://amd64.ocp.releases.ci.openshift.org/releasestream/4.16.0-0.ci/release/4.16.0-0.ci-2024-02-02-031544

[sig-trt][invariant] No new alerts should be firing expand_less 0s

{ Found alerts firing which are new or less than two weeks old, which should not be firing: PrometheusOperatorRejectedResources has no test data, this alert appears new and should not be firing}

It hit about 3-4 out of 10 on both azure and aws agg jobs. Could be a regression, could be something really rare.