Can we just focus on real problems, like stable and safe application of existing models? I'm just exhausted with the bullshit.