Artūrs Kaņepājs comments on Is There An AI Safety GiveWell?