I suppose the question is how much faith you have in oversight boards to actually produce ethical outcomes. My worry wouldn't be "they'll cut down on competiti

I suppose the question is how much faith you have in oversight boards to actually produce ethical outcomes. My worry wouldn't be "they'll cut down on competition to ensure ethical behavior", it would be "they'll block competition and protect their member instead of ensuring ethics." I think the top-level essay here is extremely cogent: oversight is almost always limited to ensuring competence. In medicine, that gets us 'ethics' when ethics means "providing good care and avoiding bad care". (It hasn't gotten us ethics when the question is anything complicated, like defining mental illness.) Competence in data science and software is largely orthogonal to ethics, so I expect an oversight board would achieve very little in that regard.